Explore the driving factors affecting cloud computing adoption in zero-trust environments using the extended TOE and TAM conceptual framework. The study addresses developed research questions using multiple regression models to investigate the correlation of the variables' relative advantage, compatibility, complexity, organizational competency, top management support, and training and education influencing cloud services' perceived ease of use (PEOU) and perceived usefulness (PU). Furthermore, it evaluates the competitive pressure and trading partner support variables' direct influence on cloud adoption intention. The study uses statistical data from a survey questionnaire measurement tool and applies the quantitative predictive correlation research design to describe and measure the variables' relationship. The data sampling consists of IT professionals with cloud computing technology and zero-trust security experience employed by government and public agencies. The findings interpret the research questions' context using correlation and regression analysis to determine the extended TOE and TAM variables' statistical significance in influencing IT professionals' perspectives of cloud services acceptance. The study contributes to the literature on cloud technology and zero trust security for future studies.
Research Article
Factors Influencing Cloud Computing Adoption in a Zero-Trust Environment
https://doi.org/10.21203/rs.3.rs-3152878/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Perceived usefulness
perceived ease of use
cloud computing security
TOE cloud adoption
TAM cloud adoption
zero trust security
Security-related factors are the significant challenge that hinders public agencies' adoption of cloud computing services [1]. The dynamic, distributed, and virtual nature of cloud computing presents unique challenges for legacy security tools to address security threats. 78% of the organizations that utilize cloud services acknowledged that the traditional security framework is inadequate for cloud service environments [2]. Security vulnerabilities, data protection, privacy concerns, and cloud providers' reliability, dependability, and honesty are significant barriers impacting cloud adoption progress [3]. Organizations must address cloud security issues to protect sensitive information assets by shifting away from traditional security strategies [4]. Over 70% of the world's businesses use cloud services, according to the Cloud Security Alliance (CSA). Significant security improvements are needed to promote greater enterprise-wide adoption of cloud services [6]. Cloud security issues pose new risks influencing the adoption process [7]. Researchers have studied the zero-trust security principle for cloud services, leading to security framework development. For instance, the National Institute of Standards and Technology (NIST) developed the zero-trust network (ZTN). Nonetheless, there are still limited studies on cloud computing services adoption in zero-trust environments based on the literature reviews of this research.
The quantitative research focuses on understanding modernizing public agencies' IT infrastructure for cloud services, addressing the technology's dynamics, distribution, and virtualization features that pose significant security challenges to traditional infrastructures [5]. The study uses the technology acceptance model (TAM) and technology-organization-environment (TOE) framework to assess factors influencing cloud computing adoption. Investigating the models' relative advantage, compatibility, complexity, organizational competency, top management support, training and education, perceived usefulness, and perceived ease of use variables to identify the significant effects affecting IT professionals' perspectives on cloud adoption intention in a zero-trust environment. Statistical data from IT professionals is analyzed using descriptive and inferential statistics to determine the correlations between the investigated variables and their outcomes [8]. The findings would contribute to the cloud body of knowledge research, providing future researchers with an understanding of cloud services and zero trust, ensuring compliance with government regulations, such as Executive Order 14028: Improving the Nation's Cybersecurity, Section 3, Modernizing Federal Government Cybersecurity.
Literature reviews supported the TOE and TAM models' relevancy for understanding cloud computing adoption. The present study uses the models to understand the factors affecting cloud computing adoption in zero-trust environments.
2.1 Technology Acceptance Model (TAM)
The TAM model emphasizes understanding the correlation between individuals and systems acceptance. Various authors' studies have used the model to understand information systems dealing with users' behavioral intentions regarding information technology [9]. The TAM model derives from the theory of reasoned action (TRA). The rationale behind the TAM's development is to provide suitability for explaining and predicting the behaviors of users toward computer use intentions [10]. The model focuses on the general determinants of computer acceptance, explaining users' behaviors toward various computing technologies. It identifies individuals' behavior concerning technology acceptance in the context of perceived usefulness and ease of use [11]. It tests the specific beliefs of perceived usefulness and perceived ease of use. The perceived usefulness reflects a person's belief that using an adopted technology would enhance their job performance. The perceived ease of use is how a person believes using the adopted technology would reduce work efforts [12, 13]. Tahar et al. [14] reveal that technology systems' ease of access impacts perceived ease of use, contributing to users' acceptance of the technology systems. When users feel a system is easy to use, it significantly influences their interest.
2.2 Technology-organization-environment (TOE) Framework
The TOE emphasizes an organization-level theory that explains the variables of technology, organization, and environment influencing adoption decisions. Tornatzky et al. [15] described the framework as a technological innovation process. The model provides broad applicability and explanatory power for explaining various adoptions of organizational systems innovations. Baker [16] concluded that in various empirical studies, TOE provides the ability to address technological innovation. Al-Mamary et al. [17], the TOE model is a holistic instrument for understanding organizational-level impacts. It provides a practical analytical framework for studying the adoption and assimilation of different IT innovation technology influencing organizations' processes.
2.3 Conceptual Framework
Substantial empirical literature reviews supported the TAM and TOE models' effectiveness in understanding technology adoption [18–21]. The two models combined serve as the theoretical and conceptual framework for examining cloud computing adoption from the perspectives of individuals and organizations [22]. Figure 1 presents the conceptual framework of the extended TAM and TOE models used to guide the present research in addressing the literature gap on the factors influencing cloud computing adoption in a zero-trust environment. The figure illustrates the conceptual links between the TAM and TOE constructs and whether it correlates with cloud adoption intention to understand the driving factors affecting cloud computing adoption in zero-trust environments.
Zero trust encompasses a new security concept created by John Kindervag in 2010 to address the security flaws in traditional perimeter models that provide access to enterprise or organization resources based on implicit trust [24]. It represents a set of security principles that shift organizations' IT security architecture away from traditional security perimeter protection [25]. Zero trust differs in the context that it does not define subjects (e.g., device, user, application) inside a predefined security perimeter as trusted. The cybersecurity paradigm moves security defenses from static network-based perimeters to focus on the users, assets, and resources regardless of network segmentation or location [26]. Alevizos et al. [27] acknowledged that zero-trust architecture protects individual devices and users beyond the traditional perimeter security model and ensures endpoints are secured across heterogeneous environments. The security concept is gaining momentum, defending against lateral movement threats in today's borderless networks. By default, Zero trust does not inherently trust any devices, systems, users, or applications based on a network location. It provides new access control policies that embrace modern technology, such as protecting scattered and distributed cloud computing resources [28]. It protects information access using continuous authentication to ensure that only authorized users interact with the system [29]. The security paradigm has no trusted network boundary (internal network) and untrusted network boundary (external networks). It strictly enforces the least privileged access control strategy by inspecting and logging all network traffic [30]. It establishes a centralized dynamic access control and management of network resources to effectively mitigate insider threats without human intervention and eliminate the traditional practice of perimeter defense [31]. Universal authentication is applied to entities where devices, users, and applications operate, regardless of being internal or external to the organization's IT infrastructure [32]. It focuses on identities, endpoints, networks, data, apps, and infrastructure providing multiple layers of defense.
Pulse surveyed 245 digital decision-makers on crucial aspects of zero-trust security for business use, and the survey reveals the followings: 63% believe the security enhances the protection of customer data; 51% believe it provides a uniform security approach; 47% believe it reduces accidental internal breaches; 47% believe it simplifies the security stack, and 46% believe it helps maintain compliance. For protection against threats, zero-trust provides 68% accidental data leakage protection, 68% malicious internal threats protection, and 64% protection to third parties working within the organization's network. Regarding organization readiness for zero-trust implementation, most organizations already have the following security components implemented in their infrastructure: 69% activity loggings, 68% identity and access management tools, 67% network segmentation, and 62% security information and event management. Further, 87% of decision-makers believed that a zero-trust security strategy simplifies their organization's security architecture [33].
3.1 Zero-Trust Standards
The NIST 800 − 207 zero trust standard is considered comprehensive and vendor-neutral for organizational use. It ensures compatibility and protection against modern attacks for cloud-based service solutions [34]. In addition to the NIST 800 − 207 zero-trust standard, the Cybersecurity and Infrastructure Security Agency described zero-trust as representing five distinct security pillars: identity, device, network, application workload, and data. As organizations' implementation of zero-trust continues to optimize, the implemented solutions increase reliance on automated processes and integrate the security pillars to create dynamic policy enforcement [35]. Prominent commercial enterprises such as Microsoft have researched zero-trust and provided descriptive details of the security pillars: identity, endpoint, networks, applications, data, infrastructure, policy enforcement, and threat protection. Microsoft researchers emphasize strengthening these security pillars from basic to advance to optimal. For example, the initial stage of the identity security pillar could involve weak password authentication. Enhancing weak password authentication in the advanced stage would involve integrating multifactor authentication to improve security. In the optimal stage, a phishing-resistant multifactor authentication could be added, providing security measures to prevent attackers from intercepting or tricking users into revealing access information [36].
The challenges organizations encounter adopting cloud service involve IT infrastructure changes and implementing critical policies to support cloud boundary applications. Security requirements become more prominent as cloud models and services use increase [35].
4.1 Cloud Models and Services
The most significant aspects that determine the level of vulnerability in a cloud computing platform are related to the deployment models of public, private, community, and hybrid clouds and the delivery services of PaaS, IaaS, and SaaS [4, 6]. Each deployment model and delivery service have unique security implications. For instance, the security challenges in the private cloud involve security implementation and management skillset requirements. For the public cloud, the security challenges involve securing the resources being leveraged across multiple cloud consumers. Resource sharing exposes organizations' applications and data to external threats. The hybrid cloud model links private clouds to external environments, creating significant risk exposure. In the SaaS delivery service, cloud providers provide applications as a service. The data is stored in the cloud provider's data center, and organizations solely rely on cloud providers for security. Organizations lack visibility to the data storage, and the risks involve insider threats, application vulnerabilities, and system availability. In the PaaS service environment, hackers could exploit control of cloud infrastructure with malware commands. The IaaS service only provides basic security requirements requiring additional protection [37].
4.2 Cloud Data and Privacy
Organizations perceive cloud data security and privacy as crucial issues [38]. Kumar and Bhatt [39] study on enhancing multi-tenancy security in cloud computing identified privacy concerns and data security as significant problems hindering the utilization of cloud computing among customers. Despite the benefits of quick deployment, low cost, pay-per-use, scalability, ubiquitous, and on-demand services, the issues of data security and privacy remain crucial aspects of cloud services adoption [40]. Rizvi et al. [41] reveal that once organizations adopt cloud services, they have limited control over significant security aspects. They mainly rely on cloud service providers to protect their data and privacy. A significant challenge in data security is preventing data leaks by privileged user access. Cloud service providers lack transparency in their internal policies, procedures, security management practices, privileged access, compliance reporting, and breach notification to investigate illegal and inappropriate activities [42]. As described by Kerman et al. [30] and Microsoft (no date), the zero-trust core security component of implicit trust leverages the least privilege, promoting the mindset that networks cannot be trusted, ensuring end-to-end communications encryption, and continuous monitoring and adjusting of network devices. These measures are appropriate security features for cloud data security challenges.
4.3 Cloud Threats and Risks
The dynamic scaling of cloud services poses numerous threats [43]. Organizations that adopt cloud services without a clear understanding of security issues are faced with the inability to identify threats promptly. Leading to data loss and not complying with regulatory protection requirements. A comprehensive understanding of cloud security protection against known and unknown risks is vital. Tabrizchi and Rafsanjani [44] classified cloud security issues as follows: policy security, user-oriented security, data storage security, application security, and network security. The security policies include the standards to prevent attacks using preventive measures, such as service-level agreements or client management issues. User-oriented security includes identification, authentication, authorization, and access management issues. Data storage security issues involve data warehouses, confidentiality, integrity, and availability (CIA) of data management. The application security includes different platforms and frameworks that pose various vulnerabilities. Network security involves cloud services dependent on network connections that could encounter threats from flooding attacks, such as denial of service (DoS), distributed denial of service (DDoS), or Internet protocol vulnerabilities.
Microsoft [45] identified the zero-trust security principle as a holistic approach to protecting identities, endpoints, networks, data, applications, and infrastructure. It is a comprehensive end-to-end strategy integrating many protection elements. Pulse's research of zero-trust protection reveals the statistical effectiveness of the security as follows: 63% of participants identified zero-trust protects customer data; 68% of participants believed zero-trust protects against accidental data leakage; 67% of participants believed zero-trust protects against malicious insider threat; 64% of participants believed zero trust protects against third-party insider threat; 56% of participants believed zero trust protects against unauthorized remote access [33]. As supported by Microsoft [45] and Pulse [33], zero trust security would be beneficial in mitigating cloud computing's threats and risks shared by Al-Shqeerat et al. [43] and Tabrizchi and Rafsanjani [44].
4.4 Cloud Trust Issues
Consumers' mistrust of cloud service providers (CSP) relates to various parameters [46]. Asadi et al. [38] emphasize that privacy and security protection determine trust in a cloud service environment. Trust is perceived as an entity to be reliable, dependable, and honest [47]. Ahmed et al. [48] define trust as a complex social phenomenon involving the factors of expectation, a specific expected behavior from the trustee based on belief and risk. Belief is the predicted behaviors resulting from the trustee's performance. Risk is the worth of exposing the organization to certain threats. Jin et al. [49], once organizations transition to cloud services, data control passes from a trusted domain to an environment where the organization lacks clear insight into security assurance. Assessing trust issues is necessary for developing positive associations and relationships with cloud providers [50]. Exploring and developing trust frameworks could help capture needed parameters to establish and manage trust efficiently. Van Der Werff et al. [3] present trust as the core barrier facing cloud adoption. It prevents organizations from exploring the benefits of cloud services. Addressing the trust problem is necessary to close the gap between consumers and cloud providers, which is vital to cloud consumers' decision-making when considering whether to adopt cloud-based services [51]. Zero-trust could strengthen consumers' trust in cloud services by providing access to organizations' resources based on implicit trust [24]. The security concept involves modifying existing IT architectures, developing real-time identity protection, identifying governance, and modernizing access management with robust mechanisms for controlling users, devices, service verification, authorization, and authentication [34, 52].
The prevalent use of cloud services brings security concerns that must be addressed to ensure regulatory compliance. Executive Order 14028 on modernizing federal government cybersecurity introduced zero-trust as a recommended security strategy. The zero-trust security has become a standard requirement for federal agencies' IT modernization when adopting cloud services [26]. Agencies are recommended to implement the National Institute of Standards and Technology (NIST) zero-trust architecture to ensure risk protection for cloud-based services. Beutel and Caron [53] acknowledged that government agencies' traditional ways of IT modernization no longer serve their purpose. The Government Accountability Office (GAO), a respected government watchdog, indicated that federal agencies continue to depend on outdated legacy IT systems and have failed to modernize IT, wasting billions of dollars.
The quantitative, predictive correlation research design was used for the study. The quantitative approach used predictive correlational statistics to describe and measure the degree of association between two or more variables [8]. The study's independent variables relating to the research topic of factors influencing cloud computing adoption in zero-trust environments were from the TAM and TOE models: relative advantage, complexity, compatibility, organizational competency, top management support, training and education, trading partner support, competitive pressure, perceived ease of use, and perceived usefulness. The dependent variables were cloud adoption intention, perceived ease of use, and perceived usefulness. Research questions were investigated using a predictive correlational research design to examine the relationships between the variables as presented. Hypotheses were developed, and a logical system was used to test the hypotheses. A deductive approach was used to test null and alternate hypotheses.
6.1 Data Collection Strategy
Pollfish, a third-party survey service, was used for the data collection using a survey questionnaire measurement tool [54]. Target respondents were filtered by demographic, career, age, and gender, as well as the requirement of survey screening questions. A priori sample size calculation determined the minimum sample size necessary to answer the study's research questions. The G*Power sample size application was used to determine the sample size to ensure an adequate sample for the research. The output consisted of input parameters that included effect size (f2 = 0.15), error probability (0.05), power (0.95), 11 predictors, and a minimum of N = 178 respondents for the data analysis collection. The data collection criteria consisted of IT professionals with experience in cloud services and zero-trust security. All the respondents (N = 178) lived in the continental United States and worked as IT professionals in government and public administration organizations. The survey included males and females aged 25 years and over. Respondents participated by rating each survey question to generate data for statistical analysis. The measurement instrument used a 5-point Likert scale rating (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree). It evaluated the relationships between the variables described in Table 1 to draw inferences on the factors driving cloud computing adoption in zero-trust environments.
Table 1. Construct Descriptions
|
Construct |
Variable type |
Level of measurement |
|
Relative advantage |
Independent |
Ordinal |
|
Compatibility |
Independent |
Ordinal |
|
Complexity |
Independent |
Ordinal |
|
Organizational competency |
Independent |
Ordinal |
|
Training & education |
Independent |
Ordinal |
|
Top management support |
Independent |
Ordinal |
|
Competitive pressure |
Independent |
Ordinal |
|
Trading partner support |
Independent |
Ordinal |
|
Perceived ease of use |
Dependent |
Ordinal |
|
Perceived usefulness |
Dependent |
Ordinal |
|
Cloud adoption intention |
Dependent |
Ordinal |
6.2 Data Analysis Strategy
The analysis collected sample data to compute statistics that estimate parameters such as variances or causal effects to define the populations [55]. The JASP SPSS software was used to perform bivariate statistics using t-tests, ANOVA, and correlation. It predicted numerical outcomes using linear and logistic regression and identified groups with cluster analysis, factor analysis, and discriminant scaling [56]. The data analysis generated from the quantitative research uncovered behaviors and trends. It did not provide insight into why respondents think, feel, or act in specific ways. It highlighted trends across data sets or study groups, not the motivation behind observed behaviors [57]. The data analysis focused on understanding the data's relationships and connection to the research questions' context. It involved multiple iterative processes that use statistical tests to interpret data [58]. Correlation and regression were used for the statistical analysis. Correlation measures the linear relationship between variables and describes the relationship's strength and direction of the two variables. Regression helps to understand the predictive power of the independent variables affecting the dependent variables if relationships exist. It illustrates the extent of change in the dependent variable's value causing the changes in the independent variables' value. The regression seeks to predict an outcome from several predictors [56].
6.3 Descriptive Statistics
The descriptive statistics summarize the research data points' trends, behaviors, and patterns. Numbers and figures illustrate the summarization of the respondents' collected data set values, such as the mean, median, mode, standard deviation, range, and skew [59]. Of the 178 participants, 100 were men, and 78 were women. Based on participants' age groups, the highest percentage of participants were aged 35 to 44 (39.88%), followed by participants aged 26 to 34 (26.40%), 45 to 54 (21.35%), and > 54 (12.36%). Participants described their level of education as postgraduate (37.64%), university (35.39%), high school (15.17%), vocational-technical college (11.24%), and middle school (0.07%). Table 2 presents the descriptive means, standard error of means, and standard deviations for the study's constructs. Cloud adoption intention had the highest mean score (4.08), while complexity had the lowest (3.46). Standard deviation scores ranged from 0.54 for relative advantage to 0.74 for competitive pressure. A Cronbach's alpha analysis was performed to measure the reliability of the survey questions. Table 3 presents the Cronbach's alpha reliability coefficients. The values for the constructs were well above 0.70, indicating the survey's reliability use for the study.
Table 2. Descriptive Statistics of Independent and Dependent Variables
|
Variable |
M |
SE |
SD |
|
Relative advantage |
3.715 |
0.040 |
0.538 |
|
Compatibility |
3.740 |
0.043 |
0.569 |
|
Complexity |
3.456 |
0.049 |
0.659 |
|
Organizational competency |
3.704 |
0.053 |
0.705 |
|
Top management support |
3.746 |
0.052 |
0.697 |
|
Training and education |
3.805 |
0.054 |
0.723 |
|
Trading partner support |
3.900 |
0.047 |
0.632 |
|
Competitive pressure |
3.789 |
0.055 |
0.740 |
|
Perceived ease of use |
3.983 |
0.054 |
0.716 |
|
Perceived usefulness |
3.885 |
0.044 |
0.582 |
|
Cloud adoption intention |
4.084 |
0.054 |
0.724 |
Note. N = 178. The variables minimum (2.0) and maximum (5.0). The information reflects the constructs' mean and standard deviation of the research dependent and independent variables.
Table 3. Cronbach's Alpha Variable Value
|
Variable |
Cronbach's α |
|
Relative advantage |
0.909 |
|
Compatibility |
0.897 |
|
Complexity |
0.919 |
|
Organizational competency |
0.900 |
|
Top management support |
0.902 |
|
Training and education |
0.904 |
|
Trading partner support |
0.903 |
|
Competitive Pressure |
0.907 |
|
Perceived ease of use |
0.903 |
|
Perceived usefulness |
0.900 |
|
Cloud adoption intention |
0.907 |
6.4 Regression Analysis
Multiple linear regression was determined to be the effective method for evaluating correlations between the dependent and independent variables. Three regression models were used to answer the study's research questions. Assumptions for each model were tested and evaluated separately before the model outputs were used to determine the hypothesis testing results. The regression models required data to be checked for normality, homoscedasticity, multicollinearity, linearity, and independence of errors [60]. The analysis used JASP, a Statistical Package for the Social Sciences (SPPS) software.
The normality assumptions for the models were evaluated using Q-Q plots for each construct. The standardized residuals were closely aligned with the regression line, indicating that normality was not violated. The homoscedasticity test, which assumed that the variance of the dependent variable was equal across different independent variable values, was also verified. At each point, the residual variation was similar across the models. Figs. 2 - 4 show a balanced random distribution of residuals around the baseline, indicating that the homoscedasticity assumption was not violated.
The linearity assumption determines whether the dependent variable is linearly related to the independent variables. Pearson's correlation coefficient was used to test the assumption, and the acceptable values range from -1 to 1. The associated correlation levels between the dependent and independent variables were classified as weak to strong. An R-value range between 0.1 to 0.3 would indicate a weak correlation. The correlation is moderate if the R-value is 0.3 to 0.5. Any R-value greater than 0.5 is considered a strong correlation. Table 4 presents the linear relationships between the dependent and independent variables.
Table 4. Pearson Correlation Summary Between the DV and IV
|
Variable Relationship |
Pearson's r |
Correlation |
|
Trading partner support -> cloud adoption intention |
0.557 |
Strong positive relationship |
|
Competitive pressure -> cloud adoption intention |
0.403 |
Moderate positive relationship |
|
Perceived ease of use -> cloud adoption intention |
0.619 |
Strong positive relationship |
|
Perceived usefulness -> cloud adoption intention |
0.625 |
Strong positive relationship |
|
Relative advantage -> perceived ease of use |
0.372 |
Weak positive relationship |
|
Compatibility -> perceived ease of use |
0.610 |
Strong positive relationship |
|
Complexity -> perceived ease of use |
0.161 |
Weak positive relationship |
|
Top management support -> perceived ease of use |
0.541 |
Strong positive relationship |
|
Training and education -> perceived ease of use |
0.534 |
Strong positive relationship |
|
Relative Advantage -> perceived usefulness |
0.486 |
Moderate positive relationship |
|
Compatibility -> perceived usefulness |
0.695 |
Strong positive relationship |
|
Complexity -> perceived usefulness |
0.265 |
Weak positive relationship |
|
Organizational competency -> perceived usefulness |
0.594 |
Strong positive relationship |
|
Top management support -> perceived usefulness |
0.579 |
Strong positive relationship |
|
Training and education -> perceived usefulness |
0.566 |
Strong positive relationship |
|
Perceived ease of use -> perceived usefulness |
0.623 |
Strong positive relationship |
Note. The correlation association between the dependent and independent variables for the multicollinearity analysis.
The multicollinearity assumption addresses high correlations between two or more independent variables that could adversely affect the use of a multiple regression model. The existence of multicollinearity could cause misleading results when determining the influence of the independent variables. Variance inflation factors (VIF) were used to test the multicollinearity assumption. Tables 5-7 present the VIF values of the multicollinearity assumption test for the models. The VIF values were not >5, indicating multicollinearity was not violated.
Table 5. VIF Values for Model 1
|
Variable |
VIF Value |
|
Relative advantage |
1.790 |
|
Compatibility |
2.675 |
|
Complexity |
1.376 |
|
Top management support |
1.911 |
|
Training and education |
1.746 |
Table 6. VIF Values for Model 2
|
Variable |
VIF Value |
|
Relative advantage |
1.801 |
|
Compatibility |
3.153 |
|
Complexity |
1.438 |
|
Organizational competency |
2.501 |
|
Top management support |
2.182 |
|
Training and education |
1.919 |
|
Perceived ease of use |
1.835 |
Table 7. VIF Values for Model 3
|
Variable |
VIF Value |
|
Trading partner support |
2.061 |
|
Perceived ease of use |
1.825 |
|
Competitive pressure |
1.579 |
|
Perceived usefulness |
2.357 |
The Durbin-Watson test evaluates the autocorrelation in the regression models' residuals to test for independence of errors. The test produces a statistic value ranging from 0 to 4. Any value that is close to 2 indicates a low autocorrelation. A value close to 0 or 4 indicates positive or negative high autocorrelation. A value equal to 2 indicates no autocorrelation. The Durbin-Watson values for Model 1 = 2.201, Model 2 = 2.068, and Model 3 = 2.199 indicate that all models' independent error assumptions were met.
Regression Model 1, Table 8, summarizes the independent variables impacting the dependent variable's perceived ease of use. R2 is the coefficient of determination that measures the dependent variable's proportion of variance explained by the independent variables. The adjusted R2 value indicates that the independent variables predicted 43.5% of the perceived ease of use variance.
Table 8. Model 1 Summary
|
Model |
R |
R² |
Adjusted R² |
|
H₁ |
0.671 |
0.451 |
0.435 |
Note. Predictors: relative advantage, compatibility, complexity, top management, training and education. Dependent variable: perceived ease of use.
The analysis of variance (ANOVA) determines the model's statistical significance. Table 9 presents the results of the ANOVA test showing the F statistic to be very significant, with a p-value of < 0.001, indicating that the model significantly predicted the independent variables influencing the dependent variable's perceived ease of use, F(5, 172) = 28.246, p < .001.
Table 9. Model 1 ANOVA Test
Note. The intercept model is omitted, as no meaningful information can be shown. Predictors: relative advantage, compatibility, complexity, top management, training and education. Dependent variable: perceived ease of use.
The regression coefficients in Table 10 present the measurement of each independent variable testing the hypotheses of Model 1's research questions (R1 – R5).
Table 10. Model 1 Regression Coefficients
|
|
|
Unstandardized Coefficients |
|
Standardized Coefficients |
|
|
|
Model |
|
B |
Standard Error |
Beta |
t |
p |
|
H₀ |
(Intercept) |
3.983 |
0.054 |
74.226 |
<0.001 |
|
|
H₁ |
(Intercept) |
0.942 |
0.322 |
2.926 |
0.004 |
|
|
|
Relative Advantage |
0.072 |
0.101 |
0.054 |
0.718 |
0.474 |
|
|
Compatibility |
0.473 |
0.116 |
0.376 |
4.068 |
<0.001 |
|
|
Complexity |
-0.174 |
0.072 |
-0.161 |
-2.422 |
0.016 |
|
|
Top Management Support |
0.215 |
0.080 |
0.209 |
2.677 |
0.008 |
|
|
Training and Education |
0.211 |
0.074 |
0.213 |
2.857 |
0.005 |
Note. Dependent variable: perceived ease of use.
Regression Model 2, Table 11, summarizes the independent variables impacting the dependent variable's perceived usefulness. R2 is the coefficient of determination that measures the dependent variable's proportion of variance explained by the independent variables. The adjusted R2 is 0.559, indicating that the independent variables explained 55.9% of the variance in perceived usefulness.
Table 11. Model 2 Summary
Note. Predictors: relative advantage, compatibility, complexity, organization competency, top management, training and education, and perceived ease of use. Dependent variable: perceived usefulness.
The ANOVA test determines the model's statistical significance. Table 12 presents the results of the ANOVA test showing the F statistic to be very significant, with a p-value of < 0.001. This result indicates that the model significantly predicted the independent variables influencing the dependent variable's perceived usefulness, F(7, 170) = 33.053, p < .001.
Table 12. Model 2 ANOVA Test
|
Model |
|
Sum of Squares |
df |
Mean Square |
F |
p |
|
H₁ |
Regression |
34.577 |
7 |
4.940 |
33.053 |
< 0.001 |
|
|
Residual |
25.405 |
170 |
0.149 |
||
|
|
Total |
59.982 |
177 |
Note. The intercept model is omitted, as no meaningful information can be shown.
The regression coefficients in Table 13 present the measurement of each independent variable testing the hypotheses of Model 2's research questions (R6 - R12).
Table 13. Model 2 Regression Coefficients
|
|
|
Unstandardized Coefficients |
|
Standardized Coefficients |
|
|
|
Model |
|
B |
Standard Error |
Beta |
t |
p |
|
H₀ |
(Intercept) |
3.885 |
0.044 |
89.047 |
<0.001 |
|
|
H₁ |
(Intercept) |
0.727 |
0.237 |
3.061 |
0.003 |
|
|
|
Relative advantage |
0.127 |
0.072 |
0.118 |
1.756 |
0.081 |
|
|
Compatibility |
0.320 |
0.091 |
0.313 |
3.533 |
<0.001 |
|
|
Complexity |
-0.055 |
0.053 |
-0.062 |
-1.035 |
0.302 |
|
|
Organizational competency |
0.062 |
0.065 |
0.075 |
0.949 |
0.344 |
|
|
Top management support |
0.073 |
0.062 |
0.087 |
1.186 |
0.237 |
|
|
Training and education |
0.101 |
0.056 |
0.126 |
1.823 |
0.070 |
|
|
Perceived ease of use |
0.198 |
0.055 |
0.244 |
3.604 |
<0.001 |
Note. Dependent variable: perceived usefulness.
Regression Model 3, Table 14, summarizes the independent variables impacting the dependent variable's cloud adoption intention. The R2 value is 0.489, and the adjusted R2 value is 0.477, indicating that the independent variables explained 47.7% of the variance in cloud computing adoption intention.
Table 14. Model 3 Summary
Note. Predictors: competitive pressure, trading partner support, perceived ease of use, perceived usefulness. Dependent variable: cloud adoption intention.
The ANOVA test determines the model's statistical significance. Table 15 presents the results showing the F statistic to be very significant, with a p-value < 0.001. This result indicates that the model significantly predicted the independent variables influencing cloud adoption intention, F(4, 173) = 41.387, p < .001.
Table 15. Model 3 ANOVA Test
Note. The intercept model is omitted, as no meaningful information can be shown. Predictors: competitive pressure, trading partner support, perceived ease of use, perceived usefulness. Dependent variable: cloud adoption intention
The regression coefficients in Table 16 present the measurement of each independent variable testing the hypotheses of Model 3's research questions (R13 – R16).
Table 16. Model 3 Regression Coefficients
|
|
|
Unstandardized Coefficients |
|
Standardized Coefficients |
|
|
|
Model |
|
B |
Standard Error |
Beta |
t |
p |
|
H₀ |
(Intercept) |
4.084 |
0.054 |
75.281 |
<0.001 |
|
|
H₁ |
(Intercept) |
0.560 |
0.287 |
1.953 |
0.052 |
|
|
|
Trading partner support |
0.183 |
0.089 |
0.160 |
2.052 |
0.042 |
|
|
Perceived ease of use |
0.352 |
0.074 |
0.348 |
4.738 |
<0.001 |
|
|
Competitive pressure |
-0.029 |
0.067 |
-0.030 |
-0.440 |
0.661 |
|
|
Perceived usefulness |
0.391 |
0.104 |
0.315 |
3.771 |
<0.001 |
Note. Dependent variable: cloud adoption intention
The research findings interpreted the variables' meaning in the context of the research questions using statistical data value to determine the hypotheses' results. Each variable's p-value determines the significance of the relationship based on p < .05. Support literature was referenced to illustrate the findings' relevancy to practitioners and the body of knowledge.
7.1 Relative Advantage
R1: Relative advantage significantly or insignificantly affect the perceived ease of use of cloud computing in a zero-trust environment.
R2: Relative advantage significantly or insignificantly affect the perceived usefulness of cloud computing in a zero-trust environment.
Relative advantage insignificantly influenced the perceived ease of use and perceived usefulness, testing the hypotheses. It relates to the technology providing significant organizational benefits and reference to the capabilities it delivers to businesses, e.g., cost performance, scalability, durability, and interoperability [23, 61]. The findings in Tables 10 and 13 suggested that IT professionals viewed relative advantage as an insignificant factor influencing perceived ease of use and perceived usefulness affecting cloud service adoption. The results differ from the findings of AlBar and Hoque [62], Behrend et al. [12], Karunagaran et al. [63], Lai [13], and Gangwar and Date [22], indicating other models may be more appropriate when studying the impact of cloud adoption in a zero-trust environment.
7.2 Compatibility
R3: Compatibility significantly or insignificantly affects the perceived ease of use of cloud computing in a zero-trust environment.
R4: Compatibility significantly or insignificantly affects the perceived usefulness of cloud computing in a zero-trust environment.
Testing the hypotheses, the results in Tables 10 and 13 indicated that IT professionals viewed compatibility as significantly affecting perceived ease of use and usefulness. It relates to the technology matching an organization's values, past experiences, existing infrastructure, and business process reducing uncertainty toward adoption [49, 62-64]. The results aligned with the TAM and TOE theoretical models, suggesting the models were appropriate when studying cloud computing in zero-trust environments.
7.3 Complexity
R5: Complexity significantly or insignificantly affects the perceived ease of use of cloud computing in a zero-trust environment.
R6: Complexity significantly or insignificantly affects the perceived usefulness of cloud computing in a zero-trust environment.
Complexity significantly predicted the dependent variable perceived ease of use and perceived usefulness based on the hypotheses results in Tables 10 and 13. It references the difficulty in understanding and adopting new technology. IT professionals viewed complexity as a significant factor affecting cloud services' perceived ease of use and perceived usefulness. Securing organizations' business processes and ensuring data privacy in a multi-tenant cloud environment raises complex issues that organizations may not have the expertise to address [62, 63]. The findings suggested that the TAM and TOE theoretical models were appropriate when studying cloud computing in zero-trust environments.
7.4 Organizational Competency
R7: Organizational competency significantly or insignificantly affects the perceived usefulness of cloud computing in a zero-trust environment.
Organizational competency insignificantly predicted the dependent variable's perceived usefulness, testing the hypothesis' result in Table 13. It refers to executives' perceptions and evaluation of the organization's awareness, commitment, resources, and established governance to adopt new technology [19, 49, 61, 64]. IT professionals viewed the effect on cloud services as insignificant. Thus, suggesting other factors may be critical to understanding cloud computing adoption when studying zero-trust environments.
7.5 Top Management Support
R8: Top management support significantly or insignificantly affects the perceived ease of use of cloud computing in a zero-trust environment.
R9: Top management support significantly or insignificantly affects the perceived usefulness of cloud computing in a zero-trust environment.
Top management support significantly predicted the dependent variables' perceived ease of use and perceived usefulness, as the hypotheses testing results indicated in Tables 10 and 13. Scholars characterized top management support as the perceptions and actions of top officials' ability to create technological innovation values for the organization [19, 62, 64, 65]. The variables' statistical relationships were significant, indicating that IT professionals viewed top management support as a significant factor affecting cloud services adoption.
7.6 Training and Education
R10: Training and education significantly or insignificantly affect the perceived ease of use of cloud computing in a zero-trust environment.
R11: Training and education significantly or insignificantly affect the perceived usefulness of cloud computing in a zero-trust environment.
The hypotheses' testing results in Tables 10 and 13 confirmed that training and education significantly predicted the dependent variables' perceived ease of use and perceived usefulness. Training and education were assumed to increase employees' familiarity with cloud technology and users' ability to use it [23]. IT professionals viewed training and education as significant factors affecting cloud services adoption.
7.7 Perceived Ease of Use
R12: Perceived ease of use significantly or insignificantly affect the perceived usefulness of cloud computing in a zero-trust environment.
R13: Perceived ease of use significantly or insignificantly affects IT professionals' cloud adoption intention in a zero-trust environment.
Perceived ease of use significantly predicted the dependent variable of perceived usefulness and cloud adoption intention based on the hypotheses' results in Tables 10 and 13. It describes the users' expectation that the targeted system is effort-free and useful in enhancing job performance [11, 13, 66]. IT professionals perceived cloud service's ease of use as a significant factor in determining the technology's usefulness in zero-trust environments. The finding aligned with Chang and Hsu [67] investigation on cloud computing benefits affecting an organization's decisions.
7.8 Competitive Pressure
R14: Competitive pressure significantly or insignificantly affects IT professionals' cloud computing adoption intentions in a zero-trust environment.
Competitive pressure insignificantly influenced cloud adoption intention based on the hypothesis result in Table 16. It addresses the pressure an organization feels from competitors to adopt a technology [23]. The assumption was that organizations that felt high levels of pressure to compete would be more likely to adopt cloud computing. However, this assumption was not supported in this study. The relationship between competitive pressure and cloud adoption intention was insignificant, indicating that IT professionals viewed competitive pressure's effects on cloud adoption intention as insignificant. Competitive pressure could positively influence cloud adoption, but federal agencies operating in zero-trust environments may not feel sufficient pressure for the influence to be significant.
7.9 Trading Partner Support
R15: Trading partner support significantly or insignificantly affects IT professionals' cloud computing adoption intentions in a zero-trust environment.
Trading partner support significantly influenced cloud adoption intention based on the hypothesis result in Table 16. It defines the effectiveness of ensuring data availability and security protection [23]. IT professionals viewed trading partner support as a significant factor affecting cloud computing adoption in a zero-trust environment.
7.10 Perceived Usefulness
R16: Perceived usefulness significantly or insignificantly affect IT professionals' cloud computing adoption intentions in a zero-trust environment.
Perceived usefulness significantly influenced cloud adoption intention based on the hypothesis result in Table 16. IT professionals viewed perceived usefulness as a critical factor when deciding whether to adopt cloud computing in zero-trust environments. The finding aligned with Tripathi and Mishra [68] study on cloud computing adoption using the TAM and TOE models.
As the study presented, cloud adoption in zero trust security environments was impacted by various factors. Future studies on cloud computing adoption in zero-trust environments should consider the limitations and implications presented in the study.
8.1 Limitations
The theoretical framework was based on the TAM and TOE models, which presented some limitations. For example, the TAM focuses on generalized external determinants affecting users' beliefs about technology usefulness and ease of use and links those beliefs to behavioral intentions to accept technology [10]. The TOE framework focuses on higher-level technological, organizational, and environmental attributes as the determinants of usefulness and ease of use [23]. These models did not specifically reflect or predict the actual behaviors of IT professionals using cloud computing in zero-trust environments. Other variables associated with different adoption and use theories may have more predictive power to explain cloud computing adoption in federal agencies. Examples of alternate models include the unified theory of acceptance and use of technology (UTAUT), the theory of reasonable action (TRA), the TAM2, and the TAM3 [13, 17, 67].
Aside from the theoretical fit model limitation, the data sampling population was restricted to IT professionals with cloud computing and zero-trust experiences. Decision-makers within organizations and agencies may have different experiences and perceptions than IT professionals who administer security systems and infrastructures. These decision-makers may have different perspectives when implementing cloud computing technologies. Thus, caution should be used when generalizing the present study's findings to all stakeholders in zero-trust environments.
Another limitation that could affect the generalizability of the study's results was the geographical focus on IT professionals working in zero-trust environments in the United States. The study targeted IT professionals in government agencies or public administration because these organizations often struggle to update their infrastructure. However, IT professionals in highly technical and innovative industries may perceive cloud computing differently. Using a sample that included IT professionals in a broader range of industries and other geographical locations could have produced different results. Thus, the study's findings should be generalized to other populations cautiously.
Another limitation was the decision to use a closed-ended online survey instrument for data collection. Creswell [8] noted that online surveys limit engagement between researchers and participants and do not allow questions and responses for clarification. Additional descriptive qualitative data could have been obtained if a mixed-methods research approach had been used. However, without this qualitative element, the study's findings remain limited to describing the statistical relationships between the selected variables.
A final limitation was the use of regression techniques to understand the variables influencing cloud intention adoption in a zero-trust environment. The regression analysis determined whether significant correlations existed between the variables, but this approach cannot indicate whether changes in an independent variable cause an effect in the dependent variables. Correlation is simply a relationship where one variable's action relates to another variable's action. Predictive correlation can be used to forecast outcomes independent variables, but predictive correlations do not equate to causation. Cause and effect relationships can only be determined using experimental or quasi-experimental methods.
8.2 Implications
Theoretical and practical implications can be drawn from the study's findings. The primary theoretical implication is that additional research is needed on cloud adoption intention in a zero-trust environment. Numerous studies supported using the TAM and TOE models to study cloud adoption intention in government agencies [18, 19, 23, 64, 69]. However, there are limited studies in zero-trust architecture environments [1, 28, 61]. This study contributed to understanding the zero-trust environment and the factors influencing cloud adoption intention in these settings. However, the findings indicate that more research is needed.
Future researchers can build on this study's findings to design future studies on cloud computing adoption in zero-trust environments. This study examined the factors influencing cloud adoption intention in such environments. However, it did not examine zero trust as an independent variable impacting the dependent variable of cloud adoption intention. Future researchers could operationalize the characteristics of zero-trust environments to determine what aspects of the environment may act as incentives or barriers to cloud adoption intentions.
Another recommendation is to use this study's findings to expand understanding of the factors influencing cloud adoption intention between zero-trust and non-zero-trust environments through alternate methodologies. Qualitatively such a study could be conducted using a multiple-case study approach with a cross-case comparison. Alternatively, quantitative researchers could evaluate the differences in adoption between the two environments using t-tests or ANOVA tests.
Lastly, the study had specific implications for researchers focused on theoretical models incorporating variables from the TAM and TOE models. The literature indicated that the TAM and TOE models were versatile enough to combine with other models to predict technology adoption [64, 69]. Several variable relationships explored in this study were insignificant, suggesting that further model development is needed to understand the most critical factors influencing cloud computing adoption intentions in zero-trust environments. A theory incorporating aspects of zero-trust environments like access control, trust, and perimeter strength would help study the organizational transitions from traditional to zero-trust infrastructures.
Competing Interests: The authors have no competing interests and received no relevant financial or non-financial interests from any organization for the submitted manuscript. All authors contributed to the study's conception and design. The author Tom Vang performed the material preparation, data collection, and analysis. The submitted manuscript has not been published elsewhere in any form or language, partially or fully.
Data Availability: Supporting data for the study's findings were conducted using a third-party survey service, Pollfish. The data could be made available from the authors upon reasonable request and with the permission of Pollfish. However, restrictions apply to the availability of these data, which were used under license for the current study and are not publicly available.
- Alassafi, M.O., AlGhamdi, R., Alshdadi, A., Al Abdulwahid, A., Bakhsh, S.T.: Determining factors pertaining to cloud security adoption framework in government organizations: an exploratory study. IEEE Access 7, 136822–136835 (2019). https://doi.org/10.1109/access.2019.2942424
- Schulze, H.: 2022 Cloud security report. Cloud Security Insider. (ISC)2; 2022. Available from: https://cyberinsiders.wpenginepowered.com/wp-content/uploads/2022/05/2022-Cloud-Security-Report-ISC2-final-ff4916d7.pdf
- Van Der Werff, L., Fox, G., Masevic, I., Emeakaroha, V.C., Morrison, J.P., Lynn, T.: Building consumer trust in the cloud: an experimental analysis of the cloud trust label approach. J. Cloud Comput. 8, 6 (2019). https://doi.org/10.1186/s13677-019-0129-8
- Mushtaq, M.F., Akram, U., Khan, I., Khan, S., Shahzad, A., Ullah, A.: Cloud computing environment and security challenges: a review. Int. J. Adv. Comput. Sci. Appl. 8, 183–195 (2017). https://doi.org/10.14569/ijacsa.2017.081025
- Cybersecurity and Infrastructure Security Agency, United States Digital Service, Federal Risk and Authorization Management Program.: Cloud security technical reference architecture [Internet]. https://www.cisa.gov/sites/default/files/2023-05/Cloud%20Security%20Technical%20Reference%20Architecture%20v2.pdf (2022)
- Ramachandra, G., Iftikhar, M., Khan, F.A.: A comprehensive survey on security in cloud computing. Procedia Comput. Sci. 110, 465–472 (2017). https://doi.org/10.1016/j.procs.2017.06.124
- Campbell, M.: Beyond zero trust: trust is a vulnerability. Computer 53, 110–113 (2020). https://doi.org/10.1109/mc.2020.3011081
- Creswell, J.W.: Research Design. 5th edn. Sage, Los Angeles (2018)
- Pramuka, B.A., Pinasti, M.: Does cloud-based accounting information system harmonize the small business needs? J. Inf. Organ. Sci. 44, 141–156 (2020). https://doi.org/10.31341/jios.44.1.6
- Silva, L.: Post-positivist review of technology acceptance model. J. Assoc. Inf. Syst. 8, 255–266 (2007). https://doi.org/10.17705/1jais.00121
- Gangwar, H., Date, H., Ramaswamy, R.: Developing a cloud-computing adoption framework. Glob. Bus. Rev. 16, 632–651 (2015). https://doi.org/10.1177/0972150915581108
- Behrend, T.S., Wiebe, E.N., London, J.E., Johnson, E.C.: Cloud computing adoption and usage in community colleges. Behav. Inf. Technol. 30, 231–240 (2011). https://doi.org/10.1080/0144929x.2010.489118
- Lai, P.C.: The literature review of technology adoption models and theories for the novelty technology. J. Inf. Syst. Technol. Manag. 14, 21–38 (2017). https://doi.org/10.4301/s1807-17752017000100002
- Tahar, A., Riyadh, H.A., Sofyani, H., Purnomo, W.E.: Perceived ease of use, perceived usefulness, perceived security and intention to use e-filing: the role of technology readiness. J. Asian Finance Econ. Bus. 7, 537–547 (2020). https://doi.org/10.13106/jafeb.2020.vol7.no9.537
- Tornatzky, L.G., Fleischer, M., Chakrabarti, A.: Processes of Technological Innovation. Lexington Books, Lexington (1990)
- Baker, J.P.: The technology–organization–environment framework. In: Dwivedi, Y., Wade, M., Schneberger, S. (eds.) Integrated Series on Information Systems, pp. 231–245. Springer Nature, New York, NY (2012)
- Al-Mamary, Y.H., Al-nashmi, M., Hassan, Y.A.G., Shamsuddin, A.: A critical review of models and theories in field of individual acceptance of technology. Int. J. Hybrid Inf. Technol. 9, 143–158 (2016). https://doi.org/10.14257/ijhit.2016.9.6.13
- Bryan, J.D., Zuva, T.: A review on TAM and TOE framework progression and how these models integrate. Adv. Sci. Technol. Eng. Syst. J. 6, 137–145 (2021). https://doi.org/10.25046/aj060316
- Gui, A., Fernando, Y., Shaharudin, M.S., Mokhtar, M., Karmawan, I.G.M., Suryanto: cloud computing adoption using TOE framework for Indonesia's micro small medium enterprises. JOIV Int. J. Inform. Vis. 4, 237 (2020). https://doi.org/10.30630/joiv.4.4.458
- Qin, X., Shi, Y., Lyu, K., Mo, Y.: Using a TAM-TOE model to explore factors of building information modelling (BIM) adoption in the construction industry. J. Civ. Eng. Manag. 26, 259–277 (2020). https://doi.org/10.3846/jcem.2020.12176
- Yim, J.S.-C., Moses, P., Azalea, A.: Predicting teachers' continuance in a virtual learning environment with psychological ownership and the TAM: a perspective from Malaysia. Educ. Technol. Res. Dev. 67, 691–709 (2019). https://doi.org/10.1007/s11423-019-09661-8
- Gangwar, H., Date, H.: Critical factors of cloud computing adoption in organizations: an empirical study. Glob. Bus. Rev. 17, 886–904 (2016). https://doi.org/10.1177/0972150916645692
- Gangwar, H., Date, H., Ramaswamy, R.: Understanding determinants of cloud computing adoption using an integrated TAM-TOE model. J. Enterp. Inf. Manag. 28, 107–130 (2015). https://doi.org/10.1108/jeim-08-2013-0065
- Teerakanok, S., Uehara, T., Inomata, A.: Migrating to zero trust architecture: reviews and challenges. Secur. Commun. Netw. 2021, 1–10 (2021). https://doi.org/10.1155/2021/9947347
- Garbis, J., Chapman, J.W.: Zero Trust Security: An Enterprise Guide. Apress, Berkeley, CA (2021)
- Rose, S., Borchert, O., Mitchell, S., Connelly, S.: Zero trust architecture [Internet]. NIST Special Publication 800-207. https://doi.org/10.6028/nist.sp.800-207 (2020)
- Alevizos, L., Ta, V.T., Eiza, M.H.: Augmenting zero trust architecture to endpoints using blockchain: a state‐of‐the‐art review. Secur. Priv. 5, e191 (2021). https://doi.org/10.1002/spy2.191
- Mandal, S., Khan, D.A., Jain, S.: Cloud-based zero trust access control policy: an approach to support work-from-home driven by COVID-19 pandemic. New Gener. Comput. 39, 599–622 (2021). https://doi.org/10.1007/s00354-021-00130-6
- Junquera-Sánchez, J., Cilleruelo, C., De-Marcos, L., Martinez-Herráiz, J.-J.: Access control beyond authentication. Secur. Commun. Netw. 2021, 1–11 (2021). https://doi.org/10.1155/2021/8146553
- Kerman et al.: [Project Description] Implementing a Zero Trust Architecture (Draft). National Cybersecurity Center of Excellence (NCCoE), p. 17 (2020). https://csrc.nist.gov/publications/detail/white-paper/2020/10/21/implementing-a-zero-trust-architecture/final
- Jin, Q., Wang, L.: Zero-trust based distributed collaborative dynamic access control scheme with deep multi-agent reinforcement learning. ICST Trans. Secur. Saf. 8, 170246 (2021). https://doi.org/10.4108/eai.25-6-2021.170246
- Uttecht: Zero Trust (ZT) concepts for federal government architectures [Internet]. Lexington, Massachusetts: Lincoln Laboratory, Massachusetts Institute of Technology; 2020. 58 p. Report No.: 1253. https://apps.dtic.mil/sti/pdfs/AD1106904.pdf (2020)
- Pulse: Zero trust security strategy adoption [Internet]. https://tinyurl.com/cm49s8mh (2021).
- Raina, K.: What is zero trust security? Principles of the zero trust model [Internet]. https://www.crowdstrike.com/cybersecurity-101/zero-trust-security/ (2021)
- Cybersecurity and Infrastructure Security Agency.: Zero trust maturity model [Internet]. https://www.cisa.gov/sites/default/files/2023-04/zero_trust_maturity_model_v2_508.pdf (2021)
- Microsoft: Evolving zero trust: how real real-world deployments and attacks are shaping the future of zero trust strategies [Internet]. https://query.prod.cms.rt.microsoft.com/cms/api/am/binary/RWJJdT (2021)
- Gupta, B.B., Badve, O.P.: Taxonomy of DoS and DDoS attacks and desirable defense mechanism in a Cloud computing environment. Neural Comput. Appl. 28, 3655–3682 (2017). https://doi.org/10.1007/s00521-016-2317-5
- Asadi, S., Nilashi, M., Husin, A.R.C., Yadegaridehkordi, E.: Customers perspectives on adoption of cloud computing in banking sector. Inf. Technol. Manag. 18, 305–330 (2017). https://doi.org/10.1007/s10799-016-0270-8
- Kumar, P., Bhatt, A.K.: Enhancing multi‐tenancy security in the cloud computing using hybrid ECC‐based data encryption approach. IET Commun. 14, 3212–3222 (2020). https://doi.org/10.1049/iet-com.2020.0255
- Farsi, M., Ali, M., Shah, R.A., Wagan, A.A., Kharabsheh, R.: Cloud computing and data security threats taxonomy: a review. J. Intell. Fuzzy Syst. 38, 2517–2527 (2020). https://doi.org/10.3233/jifs-179539
- Rizvi, S., Ryoo, J., Kissell, J., Aiken, W., Liu, Y.: A security evaluation framework for cloud security auditing. J. Supercomput. 74, 5774–5796 (2018). https://doi.org/10.1007/s11227-017-2055-1
- Bokhari, M.U., Makki, Q., Tamandani, Y.K.: A survey on cloud computing. In: Aggarwal, V., Bhatnagar, V., Mishra, D. (eds.) Advances in Intelligent Systems and Computing, pp. 149–164. Springer, Singapore (2018)
- Al-Shqeerat, K., Mohammad, H., Faiz, S., Fajraoui, H.: Cloud computing security challenges in higher educational institutions - a survey. Int. J. Comput. Appl. 161, 22–29 (2017). https://doi.org/10.5120/ijca2017913217
- Tabrizchi, H., Rafsanjani, M.K.: A survey on security challenges in cloud computing: issues, threats, and solutions. J. Supercomput. 76, 9493–9532 (2020). https://doi.org/10.1007/s11227-020-03213-1
- Microsoft: Embrace proactive security with zero trust. https://www.microsoft.com/en-us/security/business/zero-trust (no date)
- Cayirci, E., De Oliveira, A.S.: Modelling trust and risk for cloud services. J. Cloud Comput. 7, 1–16 (2018). https://doi.org/10.1186/s13677-018-0114-7
- Goode, S.: Keeping the user in the cloud: a cognitive social capital antecedent to use continuance and trust-commitment in personal cloud storage services. Behav. Inf. Technol. 38, 701–725 (2019). https://doi.org/10.1080/0144929x.2018.1551934
- Ahmed, U., Raza, I., Hussain, S.A.: Trust Evaluation in Cross-Cloud Federation. ACM Comput. Surv. 52, 1–37 (2019). https://doi.org/10.1145/3292499
- Jin, Zhou, Luo: A framework with data-centric accountability and auditability for cloud storage. The Journal of Supercomputing 74(11), 5903–5926 (2018). https://doi.org/10.1007/s11227-018-2504-5
- Spanaki, K., Gürgüç, Z., Mulligan, C., Lupu, E.: Organizational cloud security and control: a proactive approach. Inf. Technol. People 32, 516–537 (2019). https://doi.org/10.1108/itp-04-2017-0131
- Lee, L.S., Brink, W.D.: Trust in cloud-based services: a framework for consumer adoption of software as a service. J. Inf. Syst. 34, 65–85 (2020). https://doi.org/10.2308/isys-52626
- Chickowski, E.: 10 ways to avoid zero-trust failure [Internet]. [place unknown]: Dark Reading, September 20, 2021. Available from: https://www.darkreading.com/edge-slideshows/10-ways-to-avoid-zero-trust-failure
- Beutel, R., Caron, A.: Obstacles to IT modernization. Fletcher Secur. Rev. 6, 67–74 (2019)
- Pollfish: How the Pollfish methodology works [Internet]. https://resources.pollfish.com/pollfish-school/how-the-pollfish-methodology-works/ (no date)
- Zyphur, M.J., Pierides, D.C.: Is quantitative research ethical? Tools for ethically practicing, evaluating, and using quantitative research. J. Bus. Ethics 143, 1–16 (2017). https://doi.org/10.1007/s10551-017-3549-8
- Chen, H.X.: Approaches to Quantitative Research: A Guide for Dissertation Students. Oak Tree Press, Cork (2012)
- Goertzen, M.J.: Chapter 3. Introduction to quantitative research and data. Library Technology Reports, 53(4), 12-18 (2017). https://journals.ala.org/index.php/ltr/article/view/6325/8275
- Albers, M.J.: Quantitative data analysis—in the graduate curriculum. J. Tech. Writ. Commun. 47, 215–233 (2017). https://doi.org/10.1177/0047281617692067
- Spatz, C.: Exploring Statistics: Tales of Distributions. 12th edn. Outcrop, Conway, AR (2019)
- Field, A.: Discovering Statistics Using IBM SPSS Statistics. 5th edn. SAGE Publications Ltd., London (2017)
- Ali, O., Shrestha, A., Osmanaj, V., Muhammed, S.: Cloud computing technology adoption: an evaluation of key factors in local governments. Inf. Technol. People 34, 666–703 (2020). https://doi.org/10.1108/itp-03-2019-0119
- AlBar, A.M., Hoque, M.R.: Factors affecting cloud ERP adoption in Saudi Arabia: an empirical study. Inf. Dev. 35, 150–164 (2019). https://doi.org/10.1177/0266666917735677
- Karunagaran, S., Mathew, S.K., Lehner, F.: Differential cloud adoption: a comparative case study of large enterprises and SMEs in Germany. Inf. Syst. Front. 21, 861–875 (2019). https://doi.org/10.1007/s10796-017-9781-z
- Ahmed, I., Nahar, T., Urmi, S.S., Taher, K.A.: Protection of sensitive data in zero trust model. In: Proceedings of the International Conference on Computing Advancements, pp. 1–5. ACM, (2020)
- Hassan, H., Nasir, M.H.M., Khairudin, N., Adon, I.: Factors influencing cloud computing adoption in small medium enterprises. J. Inf. Commun. Technol. 16, 21–41 (2017). https://doi.org/10.32890/jict2017.16.1.2
- Gangwar, H., Date, H., Raoot, A.D.: Review on IT adoption: insights from recent technologies. J. Enterp. Inf. Manag. 27, 488–502 (2014). https://doi.org/10.1108/jeim-08-2012-0047
- Chang, Y.-W., Hsu, P.-Y.: An empirical investigation of organizations' switching intention to cloud enterprise resource planning: a cost-benefit perspective. Inf. Dev. 35, 290–302 (2019). https://doi.org/10.1177/0266666917743287
- Tripathi, S., Mishra, V.: Determinants of cloud computing adoption: a comparative study. Pac. Asia J. Assoc. Inf. Syst. 11, 3 (2019). https://doi.org/10.17705/1pais.11303
- Robles-Gomez, A., Tobarra, L., Pastor-Vargas, R., Hernandez, R., Haut, J.M.: Analyzing the users' acceptance of an IoT cloud platform using the UTAUT/TAM model. IEEE Access 9, 150004–150020 (2021). https://doi.org/10.1109/access.2021.3125497
No competing interests reported.