Chat with us, powered by LiveChat help | excelpaper.org/
+1(978)310-4246 credencewriters@gmail.com
  

TELKOMNIKA Telecommunication, Computing, Electronics and Control

Vol. 19, No. 1, February 2021, pp. 124~133

ISSN: 1693-6930, accredited First Grade by Kemenristekdikti, Decree No: 21/E/KPT/2018

DOI: 10.12928/TELKOMNIKA.v19i1.17883  124

Journal homepage: http://journal.uad.ac.id/index.php/TELKOMNIKA

Cloud computing acceptance among public sector employees

Mohd Talmizie Amron1, Roslina Ibrahim2, Nur Azaliah Abu Bakar3
1Universiti Teknologi MARA, Terengganu, Malaysia

2,3Universiti Teknologi Malaysia, Kuala Lumpur, Malaysia

Article Info ABSTRACT

Article history:

Received Jun 4, 2020

Revised Jul 20, 2020

Accepted Aug 27, 2020

Cloud computing is one of the platforms that drive organisations and users to
be better prepared for a simpler computing platform and offers significant

benefits to the quality of work. The transition from conventional computing to

the virtual world helps organisations to maximise their potential. However, not

all users can accept cloud computing adoption. Failure to understand the

factors of user’s acceptance will negatively impact the organisation’s mission

of empowering the technology. Therefore, this study proposes to assess to

what extent the users are accepting cloud computing. This study adopts the

unified theory of acceptance and use of technology (UTAUT) and six

technological and human factors assessed for the Malaysian public sectors.

Survey data from several ministries were analysed using partial least squares-

structural equation modelling (PLS-SEM). The study found out that

performance expectancy, compatibility, security, mobility, information

technology (IT) knowledge, and social influence had a significant impact on

the user’s intention to accept cloud computing. The results of this study

contribute to a clear understanding of the factors affecting the Malaysian

public sectors about cloud computing.

Keywords:

Acceptance

Behavioural intention

Cloud computing

PLS-SEM

Public sector

UTAUT

This is an open access article under the CC BY-SA license.

Corresponding Author:

Mohd Talmizie Amron

Faculty of Computer and Mathematical Sciences

Universiti Teknologi MARA, Cawangan Terengganu

21080 Kuala Terengganu, Terengganu, Malaysia

Email: [email protected]/[email protected]

1. INTRODUCTION
Recent technological advancements have brought a new dimension to the patterns of computerisation.

Previously, every organisation competed in the information communication and technology (ICT)

infrastructure with a variety of tools, devices, hardware, software, and more. However, in today’s rapidly

changing technology, with the transition to the industrial revolution 4.0 (IR 4.0) environment, it has opened a

new dimension to the world of computing. The emergence of cloud computing technology as a new platform

for computing has opened the eyes of technology industry players to further benefit from this innovation.

Many studies have proven that this technology provides many benefits to the industry and users such as its

ability to reduce operating costs, improve collaboration, more secure security levels, and more mobile

accessibility [1, 2]. Cloud computing allows more users and organisations to share resources that are optimised

for their users. This scenario will reduce user dependence on hardware and software installed on an individual

workstation.

In 2018, the Asia Cloud Computing Association (ACCA) report listed Southeast Asian countries in

cloud computing implementation. The report placed Malaysia at 8th, far behind Singapore at the top [3]. The

report stated that Malaysia had a high potential for developing cloud computing applications as Malaysia

TELKOMNIKA Telecommun Comput El Control 

Cloud computing acceptance among public sector employees (Mohd Talmizie Amron)

125

government had a clear policy and strategy to enable the delivery of cloud-based public services as well as to

drive the private sectors’ adoption of cloud technologies. However, some focus needs to be improved, such as

physical cloud infrastructure and internet speed to effectively reach the aims of the Malaysian Public Sector

ICT Strategies [4].

The Government of Malaysia has introduced a cloud-based unified communication and collaboration

services as an initiative to enhance cloud computing technology in the public sectors. This service is a platform

that integrates all communication channels such as email, live telecast calls, video conferences, instant

messaging, and big transfer files application. The centralisation of communication channels for ministries and

government agencies in the cloud is an effort to optimise the use of existing resources as well as a more

comprehensive saving effort. The implementation of the service known as MyGovUC covers all ministries and

almost all federal-level agencies since 2017.

However, according to reports of impact studies conducted by the regulatory agency on the service,

the use of this service was significantly lower than the number of account holders across Malaysia. The report

found out that 75% of users only used email applications, while only 43% used teleconferencing applications

and only half used big mail transfer applications. The study concluded that this service usage rates other than

email were due to a lack of knowledge and skills among consumers, a lack of infrastructure that could support

teleconference applications, and poor awareness of services.

Consequently, this study finds out that there is a significant problem in the adoption of cloud-based

services as its implementation cannot be matched with the usage of services and applications offered.

MAMPU’s report [5] showed that there is a gap that needs to be addressed, which is to determine factors that

influence these employees to accept cloud computing in their daily work. Then, a research model that could

evaluate the public sector employees’ acceptance of cloud-based services is developed, validated, and tested.

This study aims to identify the factors that can be considered, which can affect the cloud-based applications

used by the Malaysian public sectors. This study adopts the unified theory of acceptance and use of technology

(UTAUT) and six technological and human factors assessed for the Malaysian public sectors.

2. LITERATURE REVIEW
2.1. Cloud computing background

Cloud computing can be defined as “a model for enabling ubiquitous, convenient, on-demand network

access to a shared pool of configurable computing resources that can be rapidly provisioned and released with

minimal management effort or service provider interaction” [6]. Cloud computing is used to share resources

(data and applications) in a cloud platform that host space over the internet. There are three types of deployment

model; software as a service (SaaS), platform as a service (PaaS), and infrastructure as a service (IaaS) [6].

Besides, there are three types of service models in cloud computing; public cloud (provides services and

infrastructure to the public and organisations in a shared manner), private cloud (provides dedicated services

and support to one organisation), and hybrid cloud (a combination of public and private cloud). According to

the report by RightScale [7], a public cloud used for the business sector is 33% compared to the hybrid cloud,

which is 28% and private cloud, which is 17%.

2.2. Information technology innovation acceptance

The acceptance of new technologies by consumers or organisations varies depending on how

technology is going them perform the task faster and better [8]. The skill and magnitude of technology are tools

to facilitate the job, but in many issues and situations, its efficacy and benefits for users are subjective.

Davis [9] defined acceptance as the user’s decision on how and when to use the innovations. Thinking at the

issues which often prevent users from embracing and using technologies, many concerns need to be addressed

before users or organisations in adopting cloud computing. Multiple studies were done to assess the user’s

acceptance of cloud computing at both individual levels [10, 11] and organisational levels [12, 13].

Additionally, these studies incorporated other factors which may influence the acceptance of the user in diverse

situations. Theory of reasoned action [14], Theory of planned behavior [15], technology acceptance model [9],

diffusion of innovation [16], and task-technology fit [17] are among the theories used in the acceptance of

innovation. Meanwhile, UTAUT [18] is a great research framework incorporating eight acceptance theories

based on groundbreaking studies of individual acceptance. The UTAUT is intended to explain the user’s

intention to use an IS and subsequent behaviour. Thus, UTAUT is adopted in this study as an underlying

theoretical framework to explore the acceptability of users of cloud computing in the Malaysian public sector.

2.3. UTAUT

Venkatesh et al. [18], developed UTAUT, which clarified the user’s intention to use IS and subsequent

usage behaviour. The strength of UTAUT is that it focuses on so many models and gives an investigator a

wider view of all current models [19]. Applied research has been comprehensive on the UTAUT model. This

 ISSN: 1693-6930

TELKOMNIKA Telecommun Comput El Control, Vol. 19, No. 1, February 2021: 124 – 133

126

model offers a framework that does not only describes information technology (IT) and IS adoption but also

explains how these technologies and systems are used [20]. The UTAUT model contributes substantially to the

study of technology acceptance and uses because it can integrate different TAMs [18]. Table 1 shows the

authors, areas of study, and results of the study that used the UTAUT framework in their study. The results of

previous studies using the UTAUT framework showed different research findings, especially with various

suggestions on other factors relevant to the context of the study.

Table 1. The studies that used the UTAUT framework
Author Area of study Results/Findings

[21] RFID usage in HEI All four factors of UTAUT are significant in the study.

[22] e-Learning adoption All four factors of UTAUT are significant in the study.

[23] On-demand services
user acceptance

Using UTAUT2 and incorporated with DOI. Significant factors: personalisation,
compatibility, social influence, and perceived risk.

[24] Mobile banking

adoption

Using UTAUT2. Significant factors: Performance expectancy, effort expectancy, social

influence, habit, hedonic motivation, and perceived risk.
[25] Impact of social media

technology

Incorporated with other factors. Significant factors: utilitarian value, hedonic value and social

influence.

[26] Web-based services
acceptance

All four factors of UTAUT are significant in the study.

[27] Online banking adoption Incorporated with other factors. Significant factors: brand trust, performance expectancy,

perceived risk, initial trust.
[28] Internet banking

adoption

Incorporated with other factors. Significant factors: performance expectancy, effort

expectancy, compatibility, innovativeness, and perceived technology security.

3. RESEARCH MODEL AND HYPOTHESIS DEVELOPMENT
This study proposes a research model to understand the cloud computing acceptance trend in the

Malaysian public sectors. It is comprised of three factors that originate from the UTAUT model, namely

performance expectancy, effort expectancy, and social influence. They have also incorporated six additional

factors (compatibility, security, mobility, IT knowledge, top management support and awareness) that derive

from a related study in cloud computing acceptance. Figure 1 depicts the proposed model of the study.

Figure 1. Research proposed model

3.1. Technological factors

The technological factors focus on technological aspects linked to cloud computing, features, and

factors that make this technology acceptable. These factors include performance expectancy, effort expectancy,

compatibility, security, and mobility. Each factor is explained by the basis of the selection of those factors. The

hypothesis of these factors to the acceptance of cloud computing technology is also included in this section.

a. Performance expectancy
Performance expectancy refers to the degree to which cloud computing is used in daily work, thus

strengthening the perceptions of individuals about innovations. Venkatesh et al. [18] also stressed that the

performance expectancy, as one assumes, the program would help them to achieve job performance gains. This

aspect also helps people who embrace innovation to have clear advantages compared to others [29]. Before

using cloud computing, the positive impacts are among the most important things organisations evaluate. The

organisation is looking for lucrative returns, and much investment is proliferating. Therefore, it is appropriate

to argue that performance expectation has a positive impact on the acceptance of cloud computing. Hence, the

TELKOMNIKA Telecommun Comput El Control 

Cloud computing acceptance among public sector employees (Mohd Talmizie Amron)

127

study proposes that: H1. Performance Expectancy has a positive influence on behavioural intention to accept

cloud computing.

b. Effort expectancy
Effort expectancy described the degree of ease associated with system use [18]. User experience is

important to indicate an individual’s comfort while using technology. The invention should be useful and

helpful. However, innovation is hard to accept because it is difficult to learn, not user-friendly, and too

complex. It would not fully exploit new technology. Bozan et al. [30] addressed this component as one of the

key factors and found out that it had changed considerably among users in their relationship with behavioural

intentions. Thus, individuals with adequate expectations of effort have a more definite intention towards

accepting cloud computing. Hence, the study proposes that: H2. Effort Expectancy has a positive influence on

behavioural intention to accept cloud computing.

c. Compatibility
According to Rogers [16], compatibility refers to the degree of perceived innovation by current values,

past experiences, and desires of future adopted people. A less vague concept of a future supporter is more

compatible. Technology can either be consistent or incompatible 1) with sociocultural values or beliefs, 2) with

innovations that had previously been implemented or 3) with user requirements for innovation. Sallehudin [31]

explained that compatibility of the technological innovation with an existing infrastructure or technology either

sped up or delayed its organisational acceptance rate. In additions, it is essential to ensure that existing

infrastructure and systems are perfectly fit for the innovations to be applied so as not to be harmful after they

are used. Hence, the study proposes that: H3. Compatibility has a positive influence on behavioural intention

to accept cloud computing.

d. Security
Security aspects are critical in ensuring the protection of data and information stored in cloud

computing. Information system and technology security generally focus on protecting three main aspects of

security, namely confidentiality, integrity, and availability or known as “CIA” [32]. According to

Singh et al. [33], the cloud model based on the virtual machine environment reveals stored and shared data on

the cloud becomes vulnerable for the security breach. Therefore, it is essential for a standard security

mechanism that can be applied and implemented by all stakeholders, including service providers. Hence, the

study proposes that: H4. Security has a positive influence on behavioural intention to accept cloud computing.

e. Mobility
Mobility allows applications to use cloud computing to conveniently connect over the internet, which

is one of the technology features. Taib et al. [34] stated mobility as a ubiquitous connection which allows users

to access anytime and anywhere using the services remotely. Furthermore, mobility is the main predicted factor

for the adoption of the new mobile innovation by potential users. This should be an indicator of user acceptance

as reviewed by Saxena [35], which stressed the importance of mobility factor to implement new mobile and

electronic innovation. Hence, the study proposes that: H5. Mobility has a positive influence on behavioural

intention to accept cloud computing.

3.2. Human factors

Human factors are viewpoints in assessing human response to technology. There are four factors; IT

knowledge, top management support, social influence, and awareness. Each human factor is explained by the

basis of the selection of those factors. The hypothesis of these factors to the acceptance of cloud computing

technology is also included in this section.

a. IT Knowledge
Each individual must be armed with IT knowledge in order to accept new technology. The potential

of employees to utilise technology should be observed in order not to disrupt their adoption. According to

Sallehudin et al. [31], technology-savvy employees will lead to the adoption of IT technologies through

knowledge and innovation transformation. Competent staff can develop innovation and the need for new

technologies. This scenario provides the employee with added value in constantly seeking space to improve

work productivity. The internal expertise or IT knowledge of employees in the company is another key element.

The adoption of innovation is also expected to have an impact on IT or non-IT employees [36]. Hence, the

study proposes that: H6.IT knowledge has a positive influence on behavioural intention to accept cloud

computing.

b. Top management support
Support from top management is important for the successful implementation of innovation in the

organisation. Top management is a higher-level group managing the policies and decisions of the organisation.

A great move in the acceptance process is critical, and it will lead to the successful implementation of any

project. According to Sallehudin et al. [31], in the decision process, the power of top management serves as an

agent of transition. Top management support in cloud computing implementation can ensure employees’

 ISSN: 1693-6930

TELKOMNIKA Telecommun Comput El Control, Vol. 19, No. 1, February 2021: 124 – 133

128

efficiency and reaction. Thus, this factor refers to how the top management support affects the daily tasks of

the individual employee. Hence, the study proposes that: H7. Top Management Support has a positive influence

on behavioural intention to accept cloud computing.

c. Social influence
According to Venkatesh et al. [18], social influence refers to the degree to which a person sees what

others think about how the new system should be used. In this study, social influence tests how people are

affected by their environments by motivating them to consider and use cloud-based applications. A study by

Farah et al. [24] showed that the user relies heavily on feedback and experience from others who use new

technology in the first place. The study also showed that the most significant factor in estimating the outcome

for a user to take mobile applications is social influence. Hence, the study proposes that: H8. Social Influence

has a positive influence on behavioural intention to accept cloud computing.

d. Awareness
Awareness refers to the extent to which a person is aware of cloud-based applications services.

Implementing innovation will not work unless the innovation is used. Therefore, it is essential to provide people

with knowledge of the technologies. The use of new technology should not only be acknowledged to users but

also the opportunities to be utilised. When users realise the benefits of one technology, they are motivated and

attempt to use it. Hence, the study proposes that: H9. Awareness has a positive influence on behavioural

intention to accept cloud computing.

4. RESEARCH METHODOLOGY
Quantitative data were collected using a questionnaire, and SmartPLS 3.2 software was used to

analyse the data and validate the research model. However, before that, three activities need to be considered

to ensure that the research objectives are met. These activities include instrument development, sample

preparation, and data collection. Besides, common method bias is also emphasised in this study as it involves

single-source research.

4.1. Instrument development

Based on the seven factors included in the measurement model, the instrument was developed to

collect data. Items of measurement for each factor were adopted from previous studies and had gone through

a validity and reliability process, which were face validity and expert content validity [37].

4.2. Sample

The study population comprises of Malaysian public sectors’ employees who are using MyGovUC

application services. In order to determine the sample size, the G*Power software has been used to calculate

the minimum required sample size with effect size medium (0.15), the power needed as 0.8 and eleven

predicators. The minimum number of respondents needed is 114. The individual sample is a public sector staff

comprising various positions from various government agencies that use MyGovUC services in their daily

work. Respondents must be of those who have access to MyGovUC and are experienced in using MyGovUC.

A convenience sampling and snowball technique were employed as they helped make the survey material

easier to distribute to target groups. Convenience sampling is a type of sampling whereby recruiting the

respondents that are most easily accessible [38]. Snowball technique involves the recruitment of respondents

who in turn, recruit other respondents.

4.3. Data collection

The data collection was done via three approaches which are surveying using Google Form, emailing

the questionnaire to the public relations officer (PRO) of each ministry for the escalation within the ministry

as well as distributing the paper surveys at the ministry offices. A total of 200 paper surveys were distributed,

and 190 forms were returned (95% response rate). A total of 169 responses were received from Google form,

resulting in a total of 359 responses. This response rate is above Baruch’s [39] recommendation of between

50% and 80% for an overall survey response. Only eight responses were rejected for not meeting the set criteria

and having incomplete answers. As a result, a total of 351 valid questionnaires with a response rate of 94%

were used for further data analysis. The respondents’ profile of the study is shown in Table 2.

4.3. Common method bias

Method bias can be an issue since the data collection only involved a single source. Harman’s single

factor test was carried out to determine whether bias was on the questionnaire data. Bias occurred when

Harman’s single factor test resulted in the variance value greater than 40% [40]. In this research, Harman’s

single factor test showed that the first factor had a value of 37.10% variance (less than the 40% limit of the

TELKOMNIKA Telecommun Comput El Control 

Cloud computing acceptance among public sector employees (Mohd Talmizie Amron)

129

total variance). In this regard, the results of this test showed that the method of sampling of this study was

independent of the common bias variant.

Table 2. Total of respondents (𝑁 = 351)
Demographic Category Frequency

Gender Male 148

Female 203

Age Less than 25 55
26-35 142

36-45 111

More than 46 39
N/A 4

Academic Qualification Diploma 76

Bachelor 146

Master 74

PhD 15
Others 40

5. DATA ANALYSIS AND RESULTS
This study employs a PLS-SEM data analysis approach to measure the factors that influence a user’s

intention to accept cloud computing. The PLS-SEM is chosen in this study because its goals are to predict key

target variables, and the research model is an extension of an existing theory [41]. Therefore, predicting factors

that influence the intention of the consumer to consider cloud computing is appropriate for the objective of this

study. This study assesses multivariate normality using online Web Power tools. The analysis shows that the

p-value of Mardia’s multivariate skewness and kurtosis coefficients are less than 0.05, which confirm

multivariate non-normality data. The result is available at https://webpower.psychstat.org/models/kurtosis/

results.php?url=1367ad76416b40c8f6d4e2750c8804f2.

5.1. Measurement model

As recommended by Ramayah et al. [42], the analysis should be carried out in three main assessment

criteria, namely the internal consistency reliability, convergent validity, and discriminant validity. Table 3

shows the validity of the measurement model by measuring the loadings, composite reliability (CR), average

variance explained (AVE), and variance inflation factors (VIF). Based on [43], values in Table 3 are passed

the threshold value for all criteria. Thus, the measurement model is accepted. The list of measurement

items can be found at https://bit.ly/2O3Mb3e. The study reports the discriminant validity using the

heterotrait-monotrait (HTMT) ratio. If the HTMT value is greater than 0.85 [44], it indicates a severe issue in

discriminant validity. Table 4 shows that discriminant validity has been established as all values are less

than 0.85.

5.2. Structural model

The collinearity test is performed before the evaluation of the structural model. As indicated in

Table 2, the results of collinearity test (VIF) are lower than the threshold value of 5.0 [43]. The structural model

was assessed using the standard beta value, t-values, predictive relevance (𝑄2), and the effect sizes (𝑓 2).
Table 5 shows the assessment results of the structural model for all hypotheses.

It is shown that PER, COM, SEC, MOB, ITK, and SOC have a significant relationship with the

intention to accept cloud computing with PER (𝛽 = 0.235, 𝑡 = 3.324, 𝑝 < 0.05), COM (𝛽 = 0.123,
𝑡 = 1.731, 𝑝 < 0.05), SEC (𝛽 = −0.136, 𝑡 = 2.943, 𝑝 < 0.05), MOB (𝛽 = 0.429, 𝑡 = 8.051, 𝑝 < 0.05),
ITK (𝛽 = 0.086, 𝑡 = 1.938, 𝑝 < 0.05), and SOC (𝛽 = 0.810, 𝑡 = 2.667, 𝑝 < 0.05. Thus, H1, H3, H4, H5,
H6, and H8 are supported, while H2, H7, and H9 are rejected.

5.3. Evaluating the effect sizes

The value for the determination coefficient (R2 = 0.632) is the sum of variance in the dependent

variable structure described in the research model by all the independent variables. This study suggests that the

independent variables (PER, COM, SEC, MOB, ITK, and SOC) normally explain 63.2% of variances in

intention. As per Hair et al. [43], 𝑄2 refers to the measure of how the model and its parameter estimates and
reconstructs well-observed values. The model has predictive relevance where the value of 𝑄2 > 0. Since the
value of 𝑄2 is 0.567, the cross-validated redundancy measures indicate that the structural model for this study
has predictive relevance. The details results were presented in Table 6.

 ISSN: 1693-6930

TELKOMNIKA Telecommun Comput El Control, Vol. 19, No. 1, February 2021: 124 – 133

130

Table 3. Measurement model result
Construct # Loading (>0.5) CR (>0.7) AVE (>0.5) VIF (<0.5)

Performance Expectancy (PER) 1 0.860 0.791 0.938 3.008
2 0.913

3 0.906

4 0.877
Effort Expectancy (EFF) 1 0.880 0.776 0.945 3.534

2 0.894

3 0.864
4 0.899

5 0.867

Compatibility (COM) 1 0.851 0.795 0.939 3.105
2 0.877

3 0.922

4 0.914
Security (SEC) 1 0.844 0.713 0.952 1.855

2 0.862

3 0.871
4 0.875

5 0.818

6 0.872
7 0.791

8 0.818

Mobility (MOB) 1 0.901 0.817 0.957 1.889
2 0.899

3 0.915

4 0.895
5 0.910

IT Knowledge (ITK) 1 0.878 0.788 0.949 1.661

2 0.894
3 0.893

4 0.859

5 0.913

Top Management Support (TOP) 1 0.858 0.718 0.939 2.598

2 0.883

3 0.780
4 0.888

5 0.803

6 0.867
Social Influence (SOC) 1 0.841 0.746 0.922 3.047

2 0.888

3 0.871
4 0.854

Awareness (AWA) 1 0.835 0.663 0.887 2.312

2 0.716
3 0.862

4 0.836

Behavioural Intention (BEH) 1 0.922 0.881 0.957 1.000
2 0.949

3 0.944
Actual Use (USE) 1 0.938 0.755 0.901 –

2 0.929

3 0.722

Table 4. Discriminant validity–HTMT analy
USE AWA COM EFF ITK MOB PER BEH SEC SOC TOP

USE

AWA 0.681

COM 0.699 0.547

EFF 0.131 0.101 0.208

ITK 0.697 0.652 0.713 0.171

MOB 0.600 0.674 0.484 0.066 0.524

PER 0.603 0.517 0.680 0.207 0.758 0.457

BEH 0.633 0.588 0.743 0.148 0.823 0.484 0.819

SEC 0.422 0.555 0.439 0.171 0.534 0.423 0.517 0.597

SOC 0.614 0.727 0.609 0.126 0.619 0.499 0.588 0.678 0.640

TOP 0.453 0.599 0.525 0.165 0.525 0.465 0.555 0.601 0.605 0.818

Note: Criteria: discriminant validity is established at HTMT0.85

TELKOMNIKA Telecommun Comput El Control 

Cloud computing acceptance among public sector employees (Mohd Talmizie Amron)

131

Table 5. Path coefficient result

Hypotheses Relationship β SE t-value p-value Decision

H1 PER → BEH 0.235 0.071 3.324 0.000 Supported
H2 EFF → BEH 0.081 0.061 1.320 0.093 Unsupported

H3 COM → BEH 0.123 0.071 1.731 0.042 Supported

H4 SEC → BEH -0.136 0.046 2.943 0.002 Supported
H5 MOB → BEH 0.429 0.053 8.051 0.000 Supported

H6 ITK → BEH 0.086 0.044 1.938 0.026 Supported

H7 TOP → BEH -0.027 0.061 0.434 0.332 Unsupported
H8 SOC → BEH 0.180 0.068 2.667 0.004 Supported

H9 AWE → BEH -0.027 0.053 0.509 0.305 Unsupported
H10 BEH → USE 0.626 0.044 14.380 0.000 Supported

Table 6. Result of the effect size of each hypothesis
Path Relationship f2 Decision R2 Q2

H1 PER → BEH 0.054 Small
H3 COM → BEH 0.014 Small

H4 SEC → BEH 0.029 Small

H5 MOB → BEH 0.285 Medium

H6 ITK → BEH 0.013 Small

H8 SOC → BEH 0.031 Small

H10 BEH → USE Supported 0.632 0.567
Actual Use Supported 0.359 0.291

For the effect size, the (𝑓 2) represents the relative impact of an independent variables on a dependent
variable. As set by [45], the 𝑓 2 is measured by 0.02 representing small to medium, 0.15 represents medium to
large, and 0.35 represents large effect. The supported independent variables (PER, 𝑓 2 = 0.054; COM,
𝑓 2 = 0.014; SEC, 𝑓 2 = 0.029; ITK, 𝑓 2 = 0.013; SOC, 𝑓 2 = 0.031) have small effect size on the
dependent variable. There is only MOB (𝑓 2 = 0.285) that has a medium effect size on the dependent variable.
The R2 for behavioural intention and actual use is 0.632 and 0.359, respectively, which is acceptable.

6. DISCUSSION
Several insightful results can be summarised in this study. In the technological context, the intention

to embrace cloud computing is positively linked to four constructs (performance expectancy, compatibility,

security, and mobility). These findings underpin several previous studies [12, 31, 34]. Mobility (𝛽 = 0.429,
𝑓 2 = 0.285) is the most significant factor affecting the intention of accept cloud computing based on effect
size analysis. This may result due to the accessibility of each employee through personal devices such as

smartphones and tablets, which makes it easy for them to access the information they need. Therefore, cloud

computing needs to be further supported by service providers and organisations to ensure smoother mobile

connectivity for consumers in order to achieve more advantages by using this technology.

The following supportive result is performance expectancy with 𝛽 = 0.235 and 𝑓 2 = 0.054, which
has a positive relation to cloud computing acceptance. These findings confirm the outcomes and results of other

scholars [24, 26]. Therefore, performance expectations significantly affect the intention of accepting cloud

computing. With the optimism and positive attitude towards technology, consumers will be more comfortable

doing good work and achieve better work quality. An excellent performance will lead to overall organisational

excellence. While compatibility and security have a positive relationship with the acceptance of cloud

computing with 𝛽 = 0.123 and 𝑓 2 = 0.014, and 𝛽 = −0.136 and 𝑓 2 = 0.029, respectively, it still affects the
size of the behavioural intention. This is possible because there is still a sense of uneasiness and concern for

cloud computing capabilities compatible with the current work environment and security offered.

However, two factors (top management support and awareness) that were expected to be positive did

not occur in this study. The top management support factors showed a non-significant relationship to cloud

computing acceptance denying studies from Alharbi et al. [29]. However, this finding is consistent with studies

by Tajudeen [46]. This may be so due to employees’ feeling that their top management is not playing a role in

providing clarity and is less supportive when they are having problems.

Unfortunately, the awareness factor also does not support the hypothesis of this study. Although this

factor is highlighted in the impact report of the ICT regulatory authorities in Malaysia, this study shows that

awareness does not affect the intention of accepting cloud computing in the public sector. However, it can be

concluded that despite some non-significant factors, efforts to continue to promote and improve the use of

cloud-based applications should be continued by providing accurate disclosures and information about this

technology.

 ISSN: 1693-6930

TELKOMNIKA Telecommun Comput El Control, Vol. 19, No. 1, February 2021: 124 – 133

132

7. CONCLUSION AND FUTURE RESEARCH
This study aims to identify the factors that can be considered, which can affect the cloud-based

applications used by the Malaysian public sectors. The problems raised in the MAMPU report can be addressed

by identifying factors that influence the Malaysian public sector to accept cloud-based services. Such variables

are evaluated with the study model, and substantial test results will be taken into account in order to address

the problems associated with the implementation of cloud services.

The current study proposes a model with the UTAUT and other various factors to investigate user

behaviours towards the acceptance of cloud computing. The proposed model affects user’s intention directly

and positively. This study has identified the factors of user’s intention of cloud computing acceptance such as

performance expectancy, effort expectancy, compatibility, security, mobility, IT knowledge, top management

support, social influence, and awareness. The results of the SEM reveal that performance expectancy,

compatibility, security, and mobility have a significant influence on cloud computing acceptance. This study

has also revealed that IT knowledge and social influence are the factors that enhance user’s intention behaviour

towards the use of cloud-based application services.

For future work, it is recommended that two groups of respondents are required to complete the survey

to address the common method bias issues. Furthermore, this research has employed a quantitative data

collection approach and work will be done in the future to evaluate the model by applying both the qualitative

and quantitative approach. With these approaches, it may be helpful to provide a more in-depth explanation of

the quantitative studies results. The qualitative part may include interview sessions and verification of

quantitative research results by several experts in related fields, including CIOs of the public sector agencies.

REFERENCES
[1] S. Mokwena and C. Hlebela, “Factors affecting the adoption of software as a service in south african small medium

enterprises,” 2018 Open Innov. Conf., pp. 1-6, October 2018.

[2] M. T. M. Amron, R. Ibrahim, and S. Chuprat, “A review on cloud computing acceptance factors,” Procedia Computer
Science, vol. 124, pp. 639-646, 2017.

[3] ACCA, “Cloud Readiness Index 2018,” Asia Could Computing Association, 2018
[4] MAMPU, “The Malaysian Public Sector ICT Strategic Plan 2016-2020,” 2016.
[5] MAMPU, “Report on the presentation of findings and analysis of impact studies (in Bahasa: Laporan pembentangan

dapatan dan analisis kajian impak),” 2018. [Online]. Available: http://www.1govuc.gov.my/doc/Slaid Kajian Impak

Tech Talk.pdf. [Accessed: 27-May-2019].

[6] P. Mell and T. Grance, “Cloud Computing Definition,” Cloud Computing, 2009. [Online]. Available:
http://csrc.nist.gov/groups/SNS/cloud-computing.

[7] RightScale, “2019 State of the Cloud Report,” Flexera, 2019. [Online]. Available: https://www.rightscale.com/press-
releases/rightscale-2018-state-of-the-cloud-report. [Accessed: 03-Apr-2019].

[8] K. Krell, S. Matook, and F. Rohde, “The impact of legitimacy-based motives on IS adoption success: An institutional
theory perspective,” Inf. Manag., vol. 53, no. 6, pp. 683-697, 2016.

[9] F. D. Davis, “Perceived usefulness, perceived ease of use, and user acceptance of information technology,” MIS Q.,
vol. 13, no. 3, pp. 319-340, 1989.

[10] N. Alkhater, R. Walters, and G. Wills, “An empirical study of factors influencing cloud adoption among private
sector organisations,” Telemat. Informatics, vol. 35, no. 1, pp. 38-54, 2018.

[11] A. Goddard, et al., “The two publics and institutional theory – A study of public sector accounting in Tanzania,” Crit.
Perspect. Account., vol. 40, pp. 8-25, October 2016.

[12] M. A. Al-Sharafi, R. A. Arshah, and E. A. Abu-Shanab, “Factors influencing the continuous use of cloud computing
services in organization level,” International Conference on Advances in Image Processing, pp. 189-194, 2017.

[13] F. Mohammed, O. Ibrahim, and N. Ithnin, “Factors influencing cloud computing adoption for e-government implementation
in developing countries. Instrument development,” J. Syst. Inf. Technol., vol. 18, no. 3, pp. 297-327, 2016.

[14] M. Fishbein and I. Ajzen, “Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research,”
Addison-Wesley, 1975.

[15] I. Ajzen, “The theory of planned behavior,” Organ. Behav. Hum. Decis. Process., vol. 50, pp. 179-211, 1991.
[16] E. M. Rogers, “Diffusion of Innovations, 3rd Edition,” 3rd ed. New York: The Free Press, 1983.
[17] D. L. Goodhue, R. L. Thompson, “Task-technology fit and individual performance,” MIS Q., vol. 19, no. 2, 1995.
[18] V. Venkatesh, et al., “User acceptance of information technology: Toward a unified view,” MIS Quarterly, vol. 27,

no. 3, pp. 425-478, September 2003.

[19] A. H. Chaputula and S. Mutula, “eReadiness of public university libraries in Malawi to use mobile phones in the
provision of library and information services,” Libr. Hi Tech, vol. 36, no. 2, pp. 270-288, February 2018.

[20] C. M. Chao, “Factors determining the behavioral intention to use mobile learning: An application and extension of
the UTAUT model,” Front. Psychol., July 2019.

[21] R. J. Raja Yusof, A. Qazi, and I. Inayat, “Student real-time visualization system in classroom using RFID based on
UTAUT model,” Int. J. Inf. Learn. Technol., vol. 34, no. 3, pp. 274-288, May 2017.

[22] N. M. Patel, et al., “E-learning adoption in rural-based higher education institutions in South Africa,” Open Innov.
Conf., pp. 320-324, October 2018.

TELKOMNIKA Telecommun Comput El Control 

Cloud computing acceptance among public sector employees (Mohd Talmizie Amron)

133

[23] J. A. L. Yeap, E. H. T. Yapp, and C. Balakrishna, “User acceptance of on-demand services,” International Conference
on Research and Innovation in Information Systems, July 2017.

[24] M. F. Farah, et al., “Mobile-banking adoption: empirical evidence from the banking sector in Pakistan,” Int. J. Bank
Mark., vol. 36, no. 7, pp. 1386–1413, September 2018.

[25] H. Sharma and S. P. Pillai, “Social media technology management in college of technology in Oman an empirical
analysis,” J. Int. Educ. Bus., vol. 10, no. 2, pp. 147-163, September 2017.

[26] M. Arif, K. Ameen, and M. Rafiq, “Factors affecting student use of Web-based services: Application of UTAUT in
the Pakistani context,” Electron. Libr., vol. 36, no. 3, pp. 518-534, May 2018.

[27] Y. Zhang, et al., “Exploring trust transfer between internet enterprises and their affiliated internet-only banks: An
adoption study of internet-only banks in China,” Chinese Manag. Stud., vol. 12, no. 1, pp. 56-78, January 2018.

[28] S. Rahi and M. Abd. Ghani, “The role of UTAUT, DOI, perceived technology security and game elements in internet
banking adoption,” World J. Sci. Technol. Sustain. Dev., 2018.

[29] F. Alharbi, A. Atkins, and C. Stanier, “Understanding the determinants of cloud computing adoption in Saudi
healthcare organisations,” Complex Intell. Syst., vol. 2, no. 3, pp. 1-17, July 2016.

[30] K. Bozan, K. Parker, and B. Davey, “A closer look at the social influence construct in the UTAUT Model: An
institutional theory based approach to investigate health IT adoption patterns of the elderly,” Annu. Hawaii Int. Conf.

Syst. Sci., pp. 3105-3114, 2016.

[31] H. Sallehudin, R. C. Razak, and M. Ismail, “Factors influencing cloud computing adoption in the public sector: an
empirical analysis,” J. Entrep. Bus., vol. 3, no. 1, pp. 30-45, June 2015.

[32] I. O. for S. ISO/IEC 27002:2013, “Information technology – Security techniques – Code of practice for information
security controls,” 2013.

[33] K. P. Singh, V. Rishiwal, and P. Kumar, “Classification of data to enhance data security in cloud computing,” 3rd
International Conference On Internet of Things, pp. 1-5, February 2018.

[34] S. M. Taib, R. De Coster, and J. Nyamu, “Innovation diffusion of wearable mobile computing: Pervasive computing
perspective,” in International Conference on Information Society, pp. 97-101, October 2016.

[35] S. Saxena, “Enhancing ICT infrastructure in public services: Factors influencing mobile government (m-government)
adoption in India,” Bottom Line, vol. 30, no. 4, pp. 279-296, September 2017.

[36] H. P. Borgman, et al., “Cloudrise: Exploring cloud computing adoption and governance with the TOE framework,”
Annual Hawaii International Conference on System Sciences, 2013, pp. 4425-4435, January 2013.

[37] M. T. Amron, et al., “The validity and reliability evaluation of instruments for cloud computing acceptance study,”
in 6th IEEE International Conference on Information Management (ICIM 2020), March 2020.

[38] G. Saldanha, et al., “Data collection using questionnaires,” Research methodologies in translation studies, 2014.
[39] Y. Baruch, “Response Rate in Academic Studies,” Human Relations, vol. 52, no. 4. pp. 421-438, April 1999.
[40] B. J. Babin, M. Griffin, and J. F. Hair, “Heresies and sacred cows in scholarly marketing publications,” J. Bus. Res.,

vol. 69, no. 8, pp. 3133-3138, August 2016.

[41] C. B. Astrachan, V. K. Patel, and G. Wanzenried, “A comparative study of CB-SEM and PLS-SEM for theory
development in family firm research,” J. Fam. Bus. Strateg., vol. 5, no. 1,pp. 116-128, March 2014.

[42] T. Ramayah, et al., “Partial least squares structural equation modeling (PLS -SEM),” Partial Least Squares Structural
Equation Modeling (PLS-SEM) using SmartPLS 3.0, 2nd ed., Pearson, 2018.

[43] J. F. Hair, et al., “PLS-SEM or CB-SEM: updated guidelines on which method to use,” Int. J. Multivar. Data Anal.,
vol. 1, no. 2, January 2017.

[44] R. B. Kline, “Principles and practice of structural equation modeling,” Principles and Practice of Structural Equation
Modeling, 3rd ed., The Guilford Press, 2011.

[45] J. Cohen, “Statistical Power Analysis for the Behavioral Sciences,” 2nd ed. Lawrence Erlbaum Associates, 1988.
[46] F. P. Tajudeen, N. I. Jaafar, and S. Ainin, “Understanding the impact of social media usage among organizations,”

Inf. Manag., vol. 55, no. 3, pp. 308–321, 2018.

© 2021. This work is published under
https://creativecommons.org/licenses/by/3.0/(the “License”). Notwithstanding
the ProQuest Terms and Conditions, you may use this content in accordance

with the terms of the License.

error: Content is protected !!