Computers in Human Behavior 58 (2016) 150e157

Contents lists available at ScienceDirect

Computers in Human Behavior

journal homepage: www.elsevier.com/locate/comphumbeh

Full length article

Understanding and predicting students' intention to use mobile cloud

storage services
Ibrahim Arpaci
Gaziosmanpasa University, Department of Computer Education and Instructional Technology, Tokat, Turkey

a r t i c l e i n f o a b s t r a c t

Article history: The challenges of mobile devices such as limited bandwidth, computing, and storage have led manu-
Received 2 May 2015 facturers and service providers to develop new value-added mobile services. To address these limita-
Received in revised form tions, mobile cloud computing, which offers on-demand services including platforms, infrastructure, and
23 December 2015
software have been developed. This study attempts to build a significantly improved research framework
Accepted 28 December 2015
Available online xxx
based on the Technology Acceptance Model in order to identify factors that affect students' attitudes
toward and intentions in using mobile cloud storage services. A structural equation model was used to
assess the proposed model based on the data collected from 262 undergraduate students. Results show
Keywords:
Mobile cloud computing
that perceived usefulness, subjective norm, and trust have a significantly positive effect on the attitude,
Cloud storage which in turn is a significant predictor of behavioral intentions. The research model, which explains 82%
Ubiquity of the variance in attitudes toward using mobile cloud storage services has a strong predictive power. The
Trust findings have both theoretical and practical implications for academics, managers, and educational
institutions.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction capability to control the underlying cloud infrastructure, including

Cloud computing is a specialized distributed computing para-
digm that offers dynamically scalable computing resources, such as
computational power, storage, or business applications provisioned
as a service over the Internet (Stanoevska-Slabeva, Wozniak, &
Ristol, 2010). Similar to the Internet's layered architecture, cloud
computing has hardware, software, virtualization, and manage-
ment layers. The hardware and software layers, which include hard
drives, processors, network devices, the operating systems are
responsible for storage and processing. The virtualization layer,
which separates the logical from the physical layer, is used to
provide the essential cloud characteristics of location indepen-
dence, resource pooling, and rapid elasticity. Differing from the
traditional clienteserver network topology cloud computing is able
to offer robustness and alleviate traffic congestion. Thus, the
management layer is able to monitor traffic and implement security
mechanisms throughout the cloud (Zissis & Lekkas, 2012).
Cloud computing provides services at three different levels
including Infrastructure as a Service (IaaS), Platform as Service
(PaaS), and Software as Service (SaaS). IaaS offers users with the
E-mail address: ibrahim.arpaci@gop.edu.tr.
operating systems, network, servers, storage, and applications.
Whereas, PaaS offers users with the capability to deploy applica-
tions produced using programming languages and tools supported
by service providers on the cloud infrastructure. However, SaaS
provides users with the capability to use applications provided by
cloud service providers on the cloud infrastructure. The users do
not manage the underlying cloud infrastructure or application ca-
pabilities (Shon, Cho, Han, & Choi, 2014).
Deployment models offered by cloud providers have different
levels of security. In private and community cloud, the cloud
infrastructure is managed by the organization or a third party, and
may exist on premise or off premise. However, in public cloud the
infrastructure is made available to the general public. In hybrid
cloud, the cloud infrastructure is a composition of two or more
deployment models that remain unique entities, but are bound
together by standardized or proprietary technology that enables
data and application portability (Mell & Grance, 2009).
Cloud computing has a number of unique characteristics
including the flexibility of capabilities, the scalability of infra-
structure, broad network access, location independence, reliability,
the economies of scale, cost effectiveness, and sustainability
(Buyya, Yeo, Venugopal, Broberg, & Brandic, 2009; Shon et al.,
2014). However, it has a number of challenges such as data

http://dx.doi.org/10.1016/j.chb.2015.12.067
0747-5632/© 2015 Elsevier Ltd. All rights reserved.
 I. Arpaci / Computers in Human Behavior 58 (2016) 150e157
recovery, confidentiality, privacy, integrity, availability, reliability,
and security (Fernando, Loke, & Rahayu, 2013; Kim & Park, 2013;
Wang & Dey, 2013). These risks are similar to mobile cloud
computing, which inherits the challenges and advantages of mobile
devices. On the other hand, most of the challenges of mobile de-
vices such as limited bandwidth, computing, and storage can be
addressed by mobile cloud computing services (Dinh, Lee, Niyato, &
Wang, 2013).
In this paper, mobile cloud computing refers to an infrastructure
in which different mobile devices (i.e. smartphones, tablets, and
laptops) can access various computing resources anytime and
anywhere. One of the key functionalities of mobile cloud
computing is providing users data storage services on the cloud
where they can backup, revise, access, or share files over their
mobile devices. Some popular examples of mobile cloud storage
services are Dropbox, iCloud, SkyDrive, and Google Drive. These
services can operate across different platforms including Android,
iOS, and Blackberry and enable users synchronize their application
data such as photos, videos, music, calendars, documents, and files.
Mobile device ownership has recently experienced substantial
growth and thus the user acceptance of mobile cloud services has
emerged as a significant research domain. Among others, mobile
cloud storage services are the most closely related to students'
needs as they provide them to store, share, and effectively manage
their files and documents. To this end, the present study aims to
identify the key factors affecting students' attitudes toward and
intentions in using mobile cloud storage services. The major
contribution of this paper is to identify new factors as well as to
develop an integrated model to investigate the factors affecting the
decision to introduce cloud computing services in the educational
settings.
2. Literature review
Cloud computing has been a frequent theme of recent research
(Alamri et al., 2014; Jou & Wang, 2013; Lin, Wen, Jou, & Wu, 2014;
Navimipour, Rahmani, Navin, & Hosseinzadeh, 2015; Park & Ryoo,
2013; Stantchev, Colomo-Palacios, Soto-Acosta, & Misra, 2014;
Zhang, Ma, Wu, Ordonez de Pablos, & Wang, 2014). For example,
Shin (2015) investigated the factors that influence potential user
adoption of cloud computing. His results show that user intentions
and behaviors are largely influenced by the perceived values of
cloud services, which include access, availability, security, and
reliability. In another study, Paquette, Jaeger, and Wilson (2010)
identified possible risks associated with the governmental use of
cloud computing such as the reliability of the cloud, the continuity
of the services, security to prevent unauthorized access, safety
mechanisms, data confidentiality and privacy, and the location of
legal jurisdiction.
Yang, Sun, Zhang, and Wang (2015) investigated the factors that
influence organizational readiness to adopt SaaS. Their results
suggest that organizational users need to get prepared from tech-
nological, organizational, and environmental aspects for the
adoption of SaaS. In a similar study, Gupta, Seetharaman, and Raj
(2013) focused on the adoption of cloud computing by small and
medium enterprises (SMEs). Their results suggest that the ease of
use, convenience, security, privacy, and cost reduction are the main
factors influencing cloud usage by businesses. Moreover, their re-
sults indicate that SMEs do not want to use cloud services for
sharing data and collaborating with their stakeholders as they do
not trust these services.
Rong, Nguyen, and Jaatun (2013) reported some important se-
curity challenges in cloud computing, including resource location,
interoperability and free data movement among clouds, security,
identity management and privacy, authentication and trust, system
151
monitoring and logs, and multi-tenancy issues. Moreover, they
suggested some possible solutions for these challenges such as
service level agreements for security and holistic mechanisms for
ensuring accountability in the cloud. In another study, Marston, Li,
Bandyopadhyay, Zhang, and Ghalsasi (2011) identified the
strengths (i.e., lowering costs, immediate access to resources, and
scalability), weaknesses (i.e., the quality of service, data location,
and availability), opportunities (i.e., the effective use of information
technology and mashups), and threats (i.e., the lack of standards,
security, performance, and reliability) of cloud computing from a
business perspective.
Gonza lez-Martínez, Bote-Lorenzo, Go  mez-Sanchez, and Cano-
Parra (2015) suggested the benefits of cloud computing for
educational institutions such as the availability of online applica-
tions to support education, the flexible creation of learning envi-
ronments, support for mobile learning, computing-intensive
support for teaching, learning and evaluation, the scalability of
learning systems and applications, and cost savings. Moreover, they
reported several risks of cloud computing for educational in-
stitutions such as the lack of interoperability, performance, reli-
ability, security, privacy, licensing, and price models. In another
study, Duranti and Rogers (2012) identified some concerns
violating the trust in cloud computing such as availability, the
location of legal jurisdiction, data location, trade secrets, and data
protection. Shin (2013) investigated the acceptance of cloud
computing services by government agencies based on the Tech-
nology Acceptance Model. The results show that the usefulness and
ease of use are the significant antecedents of cloud computing. In
another study, Sultan (2010) focused on the green credentials of
cloud computing and reported that the cost advantages of cloud
computing include using less power and thereby reducing carbon
footprint.
Most studies reviewed here focus on the adoption to and use of
cloud computing services by organizations. However, there is
limited number of studies that investigate the determinants of
cloud computing services adoption in educational settings. To fill
up this gap, the present study aims to identify the key factors
affecting students' attitudes toward and intentions in using mobile
cloud storage services.
3. Research model and hypotheses
The present study proposes a research framework based on the
Technology Acceptance Model (TAM), which is one of the most
widely used adoption models in predicting user acceptance of new
technologies and systems. The TAM uses Fishbein and Ajzen's
Theory of Reasoned Action (TRA) as a theoretical basis. The TRA
posits that the adoption behavior is predicted by behavioral in-
tentions, which are also predicted by attitudes toward the behavior
and subjective norms with respect to the behavior (Fishbein &
Ajzen, 1975). However, the TAM posits that two particular beliefs,
the perceived usefulness and perceived ease of use, are the main
determinants of attitudes toward using a new technology.
In order to improve the prediction power of the TAM, this study
extended the TAM by adding perceived ubiquity, subjective norm,
perceived security, perceived privacy, and trust. More importantly,
these external factors were considered significant in predicting the
adoption of mobile cloud storage services. Fig. 1 presents the
research model to be empirically tested in this study. The model
suggests that behavioral intentions to use mobile cloud storage
services are predicted by the attitude and subjective norm towards
the behavior, while the attitude is predicted by perceived useful-
ness, trust, and subjective norm. Furthermore, perceived usefulness
is predicted by the perceived ease of use and perceived ubiquity,
while trust is predicted by perceived security and perceived
152 I. Arpaci / Computers in Human Behavior 58 (2016) 150e157
Perceived ease of use
Perceived usefulness
Perceived ubiquity
Perceived security
Trust
Perceived privacy
Subjective Norm
Fig. 1. Research model.
privacy.
3.1. Perceived usefulness
Perceived usefulness can be defined as the degree to which a
student believes that using mobile cloud storage services enhances
his or her academic performance (Davis, 1989). This variable is
related to some other constructs, including performance expec-
tancy (the Unified Theory of Acceptance and the Usage of Tech-
nology, UTAUT; Venkatesh, Morris, Davis, & Davis, 2003) and
relative advantage (the Innovation Diffusion Theory, IDT; Rogers,
2003). The main features of mobile cloud storage services such as
storing data, sharing the stored data, and data synchronization
among devices enable students effectively manage their files and
documents on mobile devices. Therefore, we hypothesize that:
H1. Perceived usefulness would positively affect students' atti-
tudes toward using mobile cloud storage services.
3.2. Perceived ease of use
Perceived ease of use can be defined as the degree to which a
student believes that using mobile cloud storage services is an
effort-free behavior (Davis, 1989). This construct is similar to the
notion of effort expectancy in the UTAUT and complexity in the IDT.
The complexity of mobile cloud storage services largely depends on
how easy it is to perform key functionalities such as storing files,
accessing stored files, and sharing the files. The easier it is to
perform these tasks, the lower the level of complexity and the
easier and quicker the perception of the advantages provided by
these services. Therefore, we hypothesize that:
H2. The perceived ease of use would positively affect perceived
usefulness.
3.3. Perceived ubiquity
Perceived ubiquity can be defined as the degree to which a
student believes that using mobile cloud storage services provide
an uninterrupted connection and access to personal data anytime
and anywhere. This construct is similar to the notion of perceived
availability in the Information Systems literature. As stated before,
mobile cloud computing inherits some important advantages of
mobile devices such as mobility, accessibility, flexibility, and
ubiquity through “always-on” connectivity. Ubiquitous connection
to personal data is one of the distinguishing characteristics of
Attitude Intention to use
mobile cloud storage services. Therefore, we hypothesize that:
H3. Perceived ubiquity would positively affect perceived
usefulness.
3.4. Trust
Trust can be defined as students' confidence in the reliability
and trustworthiness of the services offered by cloud service pro-
viders. Trust in a cloud environment depends on the privacy and
security properties of the selected deployment model. If cloud
service providers follow strict security and privacy policies, trust
may be increased, which in turn positively affects students' atti-
tudes toward using these services. Therefore, we hypothesize that:
H4. Trust would positively affect students' attitudes toward using
mobile cloud storage services.
3.5. Perceived security
Perceived security refers to the degree to which students' belief
that mobile cloud storage services are secure platforms for storing
and sharing personal data. Service providers should ensure that
their main security goals are confidentiality, integrity, and avail-
ability. Integrity refers to protecting personal data from unautho-
rized modification, deletion, or fabrication. Availability refers to the
accessibility and usability of the services and data upon demand by
an authorized entity. Confidentiality refers to the prevention of
data access by unauthorized users. The perception of a high level of
security will positively affect students' trust in mobile cloud storage
services. In line with this discussion, the following hypothesis is
formulated:
H5. Perceived security would positively affect students' trust in
mobile cloud storage services.
3.6. Perceived privacy
Perceived privacy refers to the degree to which students believe
that mobile cloud storage services are safe and protects their sen-
sitive data. When a student uses these services, there is a risk of
recording or monitoring his/her personal data. Therefore, privacy
concerns may negatively affect students' trust in mobile cloud
storage services. On the other hand, the higher the perceived pri-
vacy, the more likely mobile cloud storage services will be trusted.
Therefore, we hypothesize that:
 I. Arpaci / Computers in Human Behavior 58 (2016) 150e157 153

H6. Perceived privacy would positively affect students' trust in point Likert scale ranging from “strongly disagree” to “strongly
mobile cloud storage services. agree.”

3.7. Subjective norm 4.3. Data analysis

Subjective norm can be defined as students' perception whether 4.3.1. Instrument reliability and validity
most people who are important to them think they should or Prior to conducting the main analyses, data set was checked for
should not use mobile cloud storage services (Fishbein & Ajzen, the adequacy of factor analysis assumptions. Table 1 shows the
1975). Students who perceive greater social pressure will have suitability of the set for factor analysis. The adequacy of the data for
more positive attitudes and intentions toward using these services. factor analysis was tested with the KaisereMeyereOlkin (KMO)
This suggests that the more favorable the subjective norms about measure of sampling adequacy and Bartlett's test of sphericity
mobile cloud storage services, the more positive attitudes and in- (Bartlett, 1951; Kaiser, 1970). In addition to KMO, the results of
tentions toward using these services. Accordingly, the following Bartlett's test of sphericity verified the sampling adequacy of the
hypotheses are formulated: data set for factorability.
An exploratory factor analysis (EFA) was conducted using prin-
H7. Subjective norm would positively affect students' attitudes
cipal components extraction to test the construct validity of the
toward using mobile cloud storage services.
scale. The Bartlett's test of sphericity indicated that the measures
H8. Subjective norm would positively affect students' intentions for four latent constructs were interdependent and the KMO
to use mobile cloud storage services. measure of sampling adequacy was well above the minimally
accepted level of .50 (Leech, Barrett, & Morgan, 2005). On the basis
of a scree plot of eigenvalues, a one-factor solution appeared to be
3.8. Attitude
most appropriate for each measurement. The percentage of total
variance explained ranged from 66.87 to 87.94, which are higher
Attitude towards using a new system is defined as an in-
than acceptable value of .40 for measures with one factor (Scherer,
dividual's overall affective reaction to use the system (Davis, 1989;
Wiebe, Luther, & Adams, 1988). Moreover, each measurement item
Venkatesh et al., 2003). Students' positive feelings about mobile
has a factor loading above .73 and a communality value above .56,
cloud storage services may positively affect their behavioral in-
which are higher than acceptable value of .40 (Field, 2005). The
tentions. Therefore, the following hypothesis is formulated:
corrected item-total correlation coefficients ranged from .50 to .89,
H9. Attitude would positively affect students' intentions to use indicating moderate to high homogeneity of the items. Item anal-
mobile cloud storage services. ysis using the item discrimination index also indicated that the
measurement items can reliably discriminate the subjects.
Average variance extracted (AVE) values shown in Table 2
4. Method
exceed .50 (Hair, Anderson, Tatham, & Black, 2006), indicating
that the convergent validity for the constructs of the measurement
4.1. Survey procedure and sample
model is adequate. Finally, reliability analysis indicated that the
instrument has a strong internal consistency in that Cronbach's
Before conducting the study, the participants were instructed
alpha values ranged from .81 to .93, indicating good to very good
about how to perform basic tasks such as storing files, accessing to
internal consistency (Creswell, 2005). The results of principal
the stored files, and sharing the files using mobile cloud storage
component analysis (factor loadings and communality values), in-
services. In total, 262 undergraduate students who were selected
ternal consistency reliability measures (corrected item-total cor-
using the convenience sampling method willingly participated in
relations and Cronbach's a values) and convergent validity
the study. The participants' ages ranged from 17 to 32 years old.
measures (composite reliability and AVE values) are shown in
Majority (n ¼ 245; 93.4%) of the students were categorized in the
Table 2.
18e23 years old category. Mean age was 20.25 (SD ¼ 2.25). Of the
total, 59.2% of the respondents (n ¼ 155) were females, whereas
40.8% (n ¼ 107) were males. Moreover, 76% of the respondents were 4.3.2. Common method bias
from the first year cohort, whereas 6.1% were from the second year Harman's one-factor test was used to check common method
cohort, followed by 9.5% from the third year, and 8.4% from the bias (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). All of the
fourth year. dependent and independent variables were subjected to an
exploratory factor analysis, using unrotated principal components
4.2. Instrument factor analysis, principal component analysis with varimax rota-
tion, and principal axis analysis with varimax rotation. The results
An online survey questionnaire was designed using the ques-
tionnaire items that had been successfully used in prior studies of
Table 1
perceived usefulness and perceived ease of use (e.g., Davis, 1989;
The suitability of the data for factor analysis.
Venkatesh et al., 2003), subjective norm, attitude, and behavioral
intention (e.g., Ajzen, 1991), and perceived security, perceived pri- KMO Chi-square Sig.
García, &
vacy, perceived ubiquity, and trust (e.g., Carlos Roca, Jose Perceived usefulness .81 567.35 .001
de la Vega, 2009; Arpaci, Kilicer, & Bardakci, 2015). The in-
Jose Perceived ease of use .87 819.04 .001
Perceived ubiquity .69 296.40 .001
strument has 33 items, four items for perceived usefulness, five
Perceived security .75 520.61 .001
items for perceived ease of use, three items for perceived ubiquity, Perceived privacy .76 648.68 .001
three items for perceived security, three items for perceived pri- Trust .74 445.94 .001
vacy, four items for trust, three items for subjective norm, four Subjective norm .64 327.39 .001
items for attitude, and four items for behavioral intention. Partici- Attitude .74 480.10 .001
Intention to use .83 843.18 .001
pants were asked to indicate their level of agreement using a five-
154 I. Arpaci / Computers in Human Behavior 58 (2016) 150e157

Table 2
Reliability and validity evidence.

Construct Item Internal reliability Convergent validity

Cronbach's alpha Item-total correlation Factor loading Communality Composite reliability AVE

Perceived usefulness PU1 .87 .60 .75 .56 .88 .65

PU2 .80 .89 .81
PU3 .80 .90 .81
PU4 .73 .86 .74
Perceived ease of use PEU1 .91 .76 .85 .73 .91 .66
PEU2 .75 .85 .72
PEU3 .81 .88 .78
PEU4 .75 .84 .71
PEU5 .75 .84 .71
Perceived ubiquity PUB1 .81 .70 .88 .78 .83 .65
PUB2 .72 .89 .79
PUB3 .58 .80 .63
Perceived security PS1 .91 .81 .92 .84 .91 .77
PS2 .84 .93 .87
PS3 .80 .91 .83
Perceived privacy PP1 .93 .84 .93 .86 .93 .82
PP2 .89 .95 .91
PP3 .85 .93 .87
Trust TR1 .83 .58 .75 .57 .75 .51
TR2 .61 .78 .60
TR3 .73 .87 .75
TR4 .73 .87 .76
Subjective norm SN1 .81 .74 .90 .81 .83 .62
SN2 .75 .91 .82
SN3 .50 .73 .53
Attitude AT1 .85 .60 .76 .58 .78 .50
AT2 .68 .83 .69
AT3 .80 .90 .81
AT4 .67 .82 .67
Intention to use IU1 .93 .82 .90 .81 .94 .78
IU2 .82 .90 .82
IU3 .86 .92 .85
IU4 .83 .91 .82

suggested the presence of nine factors with eigenvalues higher Table 3

than 1.0. The nine factors together accounted for 70.02% of the total Model fit indices.

variance, while the first factor explained 45.22% of the total vari- Model Acceptable values
ance. Moreover, the confirmatory factor analysis (CFA) showed that c2 285.91
the one-factor model did not fit the data (c2/df ¼ 15.59, Goodness p value <.001 .05  p  1.00 (Hoyle, 1995)
of Fit Index (GFI) ¼ .55, Adjusted Goodness of Fit Index (AGFI) ¼ .47, c2/df 2.42 <3 (Kline, 2005)
Normed Fit Index (NFI) ¼ .64, Incremental Fit Index (IFI) ¼ .68, GFI .90 .90 (Hair, Tatham, Anderson, & Black, 2006)
AGFI .86 .80 (Marsh, Balla, & McDonald, 1988)
Standardized Root Mean Square Residual (SRMR) ¼ .38, Root Mean
SRMR .15 .10 (Kline, 2005)
Square Residuals (RMR) ¼ .36, Non-Normed Fit Index (NNFI; RMR .15 <.05 (McDonald & Moon-Ho, 2002)
TuckereLewis index) ¼ .64, Comparative Fit Index (CFI) ¼ .68, Root RMSEA .07 <.08 (Hair et al., 2006)
Mean Square Error of Approximation (RMSEA) ¼ .142). These re- NFI .91 .90 (Hair et al., 2006)
sults suggested that common method bias is not of a concern in the NNFI .93 .90 (Bentler & Bonett, 1980)
CFI .94 .90 (Bentler, 1990)
data set.
IFI .94 .90 (Bollen, 1989)

4.3.3. Measurement models

Structural equation modeling (SEM) was employed via
maximum likelihood using SPSS AMOS (version 22.0) to test the
research model. The model produced acceptable fit indices as
shown in Table 3. The results of confirmatory factor analysis
demonstrated that all scales used in this study formed adequate
measurement models and thus provided evidences for the
construct validity of the measures.
4.3.4. The structural model
Hypothesized relationships were tested through a structural
model, which was developed based on the TAM. Consistent with
the hypotheses, the results show that all proposed paths among the
latent variables are statistically significant. Fig. 2 provides results of
the SEM analysis, including the path coefficients with significance
levels along with respected error terms.
4.3.5. Hypothesis testing
A summary of the hypothesis testing results was listed below:
H1. Perceived usefulness would positively affect students' atti-
tudes toward using mobile cloud storage services (b ¼ .61;
t ¼ 8.19; p < .001).
H2. Perceived ease of use would positively affect perceived
usefulness (b ¼ .50; t ¼ 9.05; p < .001).
H3. Perceived ubiquity would positively affect perceived use-
fulness (b ¼ .69; t ¼ 9.00; p < .001).
H4. Trust would positively affect students' attitudes toward
using mobile cloud storage services (b ¼ .63; t ¼ 8.52; p < .001).
H5. Perceived security would positively affect students' trust in
mobile cloud storage services (b ¼ .36; t ¼ 6.64; p < .001).
 I. Arpaci / Computers in Human Behavior 58 (2016) 150e157
R2 = .72, e = .12
Perceived ease of use
Perceived usefulness
Perceived ubiquity
R2 = .52, e = .22
Perceived security
Trust
Perceived privacy
Subjective Norm
*p < .001, Chi-Square = 285.91, df = 118, Chi-Square/df = 2.42, P-value = .000, GFI = .90, RMSEA = .074
Fig. 2. Causal model of the students' intention to use mobile cloud storage services.
H6. Perceived privacy would positively affect students' trust in
mobile cloud storage services (b ¼ .63; t ¼ 11.18; p < .001).
H7. Subjective norm would positively affect students' attitudes
toward using mobile cloud storage services (b ¼ .25; t ¼ 3.75;
p < .001).
H8. Subjective norm would positively affect students' intentions
to use mobile cloud storage services (b ¼ .35; t ¼ 4.79; p < .001).
H9. Attitude would positively affect students' intentions to use
mobile cloud storage services (b ¼ .45; t ¼ 6.10; p < .001).
5. Discussion and conclusion
Recently, mobile device ownership has experienced substantial
growth and mobile cloud services have emerged as a practical so-
lution to address the challenges of these devices such as limited
bandwidth, computing, and storage. Mobile cloud storage services
allow users to store data, share the stored data, and access the data
synchronized among devices. Although the latent customers of
mobile cloud storage services are social workers or business firms,
the adoption of these services by educational institutions and
students is a significant research area that needs investigation. As
students manage their files and documents on their mobile devices,
these services can be used by students for educational purposes,
especially for team projects in which members are working
simultaneously on the same document.
This study extended the TAM by adding new factors including
perceived ubiquity, perceived security, perceived privacy, subjec-
tive norm, and trust and then examined how these factors affected
students' attitudes toward and intentions in using mobile cloud
storage services. More importantly, this study integrated these
factors with the TAM and developed a new research model to
predict the adoption of mobile cloud storage services by confirming
the validity of the proposed model via SEM.
Results show that perceived usefulness, trust, and subjective
norm have significant effects on attitudes toward mobile cloud
storage services, whereas 82% of the variance in attitude was found
to be explainable by the combination of these factors. The high
proportion of the total variance explained indicates that the model
includes a significant portion of factors that might affect attitudes.
Perceived ease of use and perceived ubiquity emerged as mean-
ingful predictors of perceived usefulness by explaining 72% of its
variance. The combinatory effects of perceived security and
perceived privacy explained 52% of the variance in trust. The results
also suggest that attitude and subjective norm have significant ef-
fects on behavioral intentions, whereas 40% of the variance in
155
R2 = .82, e = .05 R2 = .40, e = .24
Attitude Intention to use
behavioral intentions was found to be explainable by the combi-
nation of these factors. All these findings suggest that the proposed
model successfully establishes valid links between the key factors
of the services (i.e., perceived ubiquity, perceived privacy, perceived
security, and trust) and the traditional constructs from the TAM
(i.e., perceived usefulness, attitude, intention to use), thereby
extending adoption theories on mobile cloud storage services.
In accordance with prior studies that revealed the positive ef-
fects of perceived privacy and perceived security on attitudes to-
ward cloud storage services (e.g., Lian, Yen, & Wang, 2014; Park &
Kim, 2014; Yang & Lin, 2015), the findings add to the existing
literature in that these factors serve as influential determinants for
trust in cloud storage services. The implication is that the enhanced
security of and user privacy with a mobile cloud storage service are
critical to the failure or success of the service, encouraging service
providers to invest in developing safe, reliable, and trustworthy
infrastructures and platforms that protect users' sensitive data from
unauthorized modification, deletion, or fabrication.
The results show that the perceived ease of use and perceived
ubiquity have significant effects on perceived usefulness, which in
turn positively affects attitudes. This suggests that students believe
that mobile cloud storage services would increase their produc-
tivity, enhance their effectiveness, improve their academic perfor-
mance, and ease their tasks. The results also suggest that students
find it easy to become skillful at using mobile cloud storage services
and believe that their interaction with these services would be clear
and understandable. Moreover, they believe that mobile cloud
storage services provide an uninterrupted connection and access to
personal data anytime and anywhere.
5.1. Research implications
The present study has important research implications. First, the
results suggested that the research model explains the variance of
the dependent variable better than the TAM. For example, the
perceived ease of use and perceived usefulness of the TAM explains
62% of variance in attitudes, while the research model explains 82%
of variance in attitudes toward using mobile cloud storage services.
Second, the TAM does not explicitly include any social variables.
However, subjective norm in the proposed model may capture
unique variance in attitudes and intentions. Finally, the TAM pro-
vides very general information on students' opinions about mobile
cloud storage services. However, the research model delivers more
specific information by including variables such as perceived
ubiquity, perceived security, perceived privacy, and trust. Therefore,
156 I. Arpaci / Computers in Human Behavior 58 (2016) 150e157
the research model is more likely to capture situation-specific
factors.
There are some interesting findings in this study that are
different from the previous studies investigating mobile or cloud
services. First, the findings suggest that perceived security and
perceived privacy are directly related to trust in mobile cloud
storage services and indirectly related to attitudes and behavioral
intentions toward using such services. Second, the findings show
that perceived ubiquity, which is one of the unique and important
features of mobile devices, along with the perceived ease of use is
directly related to the perceived usefulness of these services.
5.2. Practical implications
The findings have several practical implications for academics,
managers, and educational institutions. First, mobile cloud storage
services provide students a free space to store, access, and share
files and documents. This implies that academics can promote the
effective use of these services for group projects in which students
work together on a document. It is important to note that in-
stitutions with greater management support would be more likely
to adopt such services. Therefore, managers should support the use
of mobile cloud storage services to facilitate collaboration among
faculty members. Finally, educational institutions may more
strongly justify the promotion of these services among students
and academics at large. Thus, universities should have policy
guidelines on the use of such services and provide similar services
working in an intranet with secure access.
5.3. Limitations of the study
There are several issues that should be taken into consideration
in future research on related topics. First, individual differences of
the respondents were not considered in this study. The individual
factors (i.e. age, gender) may have significant effects on user atti-
tude towards and intention to use mobile cloud storage services.
Given that the respondents were recruited from Turkey, a collec-
tivist society, users from the Western societies that have individ-
ualistic cultures may lead to different adoption patterns. Future
studies may consider investigating the potential moderating effects
of these cultural differences and employ diverse samples for greater
generalizability of the proposed model. In addition, it may be useful
to include qualitative data for a deeper investigation of the key
factors affecting attitudes and intentions toward using such ser-
vices. Thus, a follow-up analysis on the indirect and direct re-
lationships among these factors is recommended. Finally, items
with high modification indices were deleted to improve the model
fit. Therefore, it would be appropriate to perform a cross-validation
with a new data sample. While the current study contributes to a
more systematic understanding of mobile cloud storage services,
future studies may extend and refine the findings by addressing
these limitations.
References
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human
Decision Processes, 50(2), 179e211.
Alamri, A., Hassan, M. M., Hossain, M. A., Al-Qurishi, M., Aldukhayyil, Y., &
Hossain, M. S. (2014). Evaluating the impact of a cloud-based serious game on
obese people. Computers in Human Behavior, 30(1), 468e475.
Arpaci, I., Kilicer, K., & Bardakci, S. (2015). Effects of security and privacy concerns on
educational use of cloud services. Computers in Human Behavior, 45(4), 93e98.
Bartlett, M. S. (1951). The effect of standardization on a c2 approximation in factor
analysis. Biometrika, 38(3/4), 337e344.
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological
Bulletin, 107, 238e246.
Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness-of-fit in the
analysis of covariance structures. Psychological Bulletin, 88, 588e606.
Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley.
Buyya, R., Yeo, C. S., Venugopal, S., Broberg, J., & Brandic, I. (2009). Cloud computing
and emerging IT platforms: vision, hype, and reality for delivering computing as
the 5th utility. Future Generation Computer Systems, 25(6), 599e616.
Carlos Roca, J., Jose  García, J., & Jose  de la Vega, J. (2009). The importance of
perceived trust, security and privacy in online trading systems. Information
Management & Computer Security, 17(2), 96e113.
Creswell, J. W. (2005). Educational research: Planning conducting and evaluating
quantitative and qualitative approaches to research (2nd ed.). Upper Saddle River,
NJ: Merrill/Pearson Education.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance
of information technology. MIS Quarterly, 13(3), 319e340.
Dinh, H. T., Lee, C., Niyato, D., & Wang, P. (2013). A survey of mobile cloud
computing: architecture, applications, and approaches. Wireless Communica-
tions and Mobile Computing, 13(18), 1587e1611.
Duranti, L., & Rogers, C. (2012). Trust in digital records: an increasingly cloudy legal
area. Computer Law & Security Review, 28(5), 522e531.
Fernando, N., Loke, S. W., & Rahayu, W. (2013). Mobile cloud computing: a survey.
Future Generation Computer Systems, 29(1), 84e106.
Field, A. (2005). Discovering statistics using SPSS. Thousand Oaks, CA: SAGE.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An intro-
duction to theory and research. Reading, MA: Addison-Wesley.
Gonz alez-Martínez, J. A., Bote-Lorenzo, M. L., Go mez-Sanchez, E., & Cano-Parra, R.
(2015). Cloud computing and education: a state-of-the-art survey. Computers &
Education, 80, 132e151.
Gupta, P., Seetharaman, A., & Raj, J. R. (2013). The usage and adoption of cloud
computing by small and medium businesses. International Journal of Informa-
tion Management, 33(5), 861e874.
Hair, J. F., Tatham, R. L., Anderson, R. E., & Black, W. (2006). Multivariate data analysis
(6th ed.). Upper Saddle River, NJ: Pearson Prentice Hall.
Hoyle, R. H. (1995). The structural equation modeling approach: basic concepts and
fundamental issues. In R. H. Hoyle (Ed.), Structural equation modeling: Concepts,
issues, and applications (pp. 1e15). Thousand Oaks, CA: Sage Publications, Inc.
Jou, M., & Wang, J. (2013). Observations of achievement and motivation in using
cloud computing driven CAD: comparison of college students with high school
and vocational high school backgrounds. Computers in Human Behavior, 29(2),
364e369.
Kaiser, H. F. (1970). A second generation little jiffy. Psychometrika, 35(4), 401e415.
Kim, M., & Park, S. O. (2013). Trust management on user behavioral patterns for a
mobile cloud computing. Cluster Computing, 16(4), 725e731.
Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.).
New York: The Guilford Press.
Leech, N. L., Barrett, K. C., & Morgan, G. A. (2005). SPSS for intermediate statistics: Use
and interpretation (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.
Lian, J. W., Yen, D. C., & Wang, Y. T. (2014). An exploratory study to understand the
critical factors affecting the decision to adopt cloud computing in Taiwan
hospital. International Journal of Information Management, 34(1), 28e36.
Lin, Y. T., Wen, M. L., Jou, M., & Wu, D. W. (2014). A cloud-based learning envi-
ronment for developing student reflection abilities. Computers in Human
Behavior, 32(3), 244e252.
Marsh, H. W., Balla, J. R., & McDonald, R. P. (1988). Goodness-of-fit indexes in
confirmatory factor analysis: the effect of sample size. Psychological Bulletin,
103(3), 391e410.
Marston, S., Li, Z., Bandyopadhyay, S., Zhang, J., & Ghalsasi, A. (2011). Cloud
computing-the business perspective. Decision Support Systems, 51(1), 176e189.
McDonald, R. P., & Moon-Ho, R. H. (2002). Principles and practice in reporting
structural equation analysis. Psychological Methods, 7(1), 64e82.
Mell, P., & Grance, T. (2009). The NIST definition of cloud computing. National
Institute of Standards and Technology, 53(6), 50.
Navimipour, N. J., Rahmani, A. M., Navin, A. H., & Hosseinzadeh, M. (2015). Expert
cloud: a cloud-based framework to share the knowledge and skills of human
resources. Computers in Human Behavior, 46(5), 57e74.
Paquette, S., Jaeger, P. T., & Wilson, S. C. (2010). Identifying the security risks
associated with governmental use of cloud computing. Government Information
Quarterly, 27(3), 245e253.
Park, E., & Kim, K. J. (2014). An integrated adoption model of mobile cloud services:
exploration of key determinants and extension of technology acceptance
model. Telematics and Informatics, 31(3), 376e385.
Park, S. C., & Ryoo, S. Y. (2013). An empirical investigation of end-users’ switching
toward cloud computing: a two factor theory perspective. Computers in Human
Behavior, 29(1), 160e170.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common
method biases in behavioral research: a critical review of the literature and
recommended remedies. Journal of Applied Psychology, 88(5), 879.
Rogers, E. M. (2003). Diffusion of innovations (5th ed.). New York: Free Press.
Rong, C., Nguyen, S. T., & Jaatun, M. G. (2013). Beyond lightning: a survey on security
challenges in cloud computing. Computers & Electrical Engineering, 39(1),
47e54.
Scherer, R. F., Wiebe, F. A., Luther, D. C., & Adams, J. S. (1988). Dimensionality of
coping: factor stability using the ways of coping questionnaire. Psychological
Reports, 62, 763e770.
Shin, D. H. (2013). User centric cloud service model in public sectors: policy im-
plications of cloud services. Government Information Quarterly, 30(2), 194e203.
Shin, D. (2015). Beyond user experience of cloud service: implication for value
sensitive approach. Telematics and Informatics, 32(1), 33e44.
 I. Arpaci / Computers in Human Behavior 58 (2016) 150e157
Shon, T., Cho, J., Han, K., & Choi, H. (2014). Toward advanced mobile cloud
computing for the internet of things: current issues and future direction. Mobile
Networks and Applications, 19(3), 404e413.
Stanoevska-Slabeva, K., Wozniak, T., & Ristol, S. (2010). Grid and cloud computing: A
business perspective on technology and applications. Berlin, Heidelberg: Springer-
Verlag.
Stantchev, V., Colomo-Palacios, R., Soto-Acosta, P., & Misra, S. (2014). Learning
management systems and cloud file hosting services: a study on students'
acceptance. Computers in Human Behavior, 31(1), 612e619.
Sultan, N. (2010). Cloud computing for education: a new dawn? International
Journal of Information Management, 30(2), 109e116.
Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of
information technology: toward a unified view. MIS Quarterly, 27(3), 425e478.
157
Wang, S., & Dey, S. (2013). Adaptive mobile cloud computing to enable rich mobile
multimedia applications. IEEE Transactions on Multimedia, 15(4), 870e883.
Yang, H. L., & Lin, S. L. (2015). User continuance intention to use cloud storage
service. Computers in Human Behavior, 52, 219e232.
Yang, Z., Sun, J., Zhang, Y., & Wang, Y. (2015). Understanding SaaS adoption from the
perspective of organizational users: a tripod readiness model. Computers in
Human Behavior, 45(4), 254e264.
Zhang, X., Ma, H., Wu, Y., Ordonez de Pablos, P., & Wang, W. (2014). Applying cloud
computing technologies to upgrade the resource configuration of laboratory
course: the case of quality engineering education platform. International Journal
of Engineering Education, 30(3), 596e602.
Zissis, D., & Lekkas, D. (2012). Addressing cloud computing security issues. Future
Generation Computer Systems, 28(3), 583e592.