Role of perceived risk in determining consumer acceptance of mobile payment: An empirical study in Vietnam
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
International Conference on Marketing in the Connected Age (MICA-2018), October 6th, 2018
Danang City, Vietnam
Role of Perceived Risk in Determining Consumer Acceptance
of Mobile Payment: An Empirical Study in Vietnam
Nhan Tran Danha*, Ha Tran Thi Phuongb
aFaculty of E-Commerce, University of Economics – The University of Danang, Danang 59000, Vietnam
bFaculty of Marketing, University of Economics – The University of Danang, Danang 59000, Vietnam
A B S T R A C T
Mobile payment is getting emergent consideration worldwide with enormous potential to explore. This study
aims to investigate the impacts of various factors on behavior intention and actual usage of mobile payment
based on a risk-extended model of UTAUT2 as well as explore impacts of various individual risk aspects on
the overall perceived risk toward mobile payment adoption. The research model was empirically examined
utilizing 329 responses conducted from Vietnamese young people. The research results confirmed that
performance expectancy, facilitating conditions, price value and habit altogether have direct effects on the
intention to adopt mobile payment. To determine usage behavior of mobile payment, the behavior intention
along with facilitating conditions and habit make impacts on the actual use. Regarding to perceived risk, it
does have negative indirect, not direct, impact on behaviors intention through the mediation of performance
expectancy. When perceived risk is included in the model, the effect of effort expectancy on behavior
intention would be diminished due to the mediation role of perceived risk between the two constructs.
Furthermore, in case of mobile payment, it is suggested that the overall perceived risk would be significantly
influenced by perceived psychological risk, time risk, performance risk, and privacy risk.
Keywords: mobile payment; UTAUT, perceived risk, technology acceptance
1. Introduction
Mobile payment is defined as “a process in which at least one phase of the payment transaction is conducted
using a mobile device (such as mobile phone, smartphone, PDA, or any wireless enabled device) capable of
securely processing a financial transaction over a mobile network, or via various wireless technologies (NFC,
Bluetooth, RFID, etc.)” [1]. Paying for transactions via mobile devices let consumers get rid of the requirement
of cash using, provide convenient and fast payment procedure, enable secure information transferring between
devices with various transaction categories ranged from individual to high scale of payment volume [2-4].
Across the global markets, growth of mobile payment is being expeditious and its potentials are realizing by
more and more commercial entities [5, 6]. According to Statista Corporation [7], the world-wide returns for
mobile payment is estimated to exceed 1 trillion U.S. dollars in 2019 due to the spread of mobile devices like
smartphones and tablets. However, mobile payment is still considered as a comparatively new research area in
comparison with other technology adoption research fields such as e-commerce, Internet banking or mobile
banking. In few recent research that issued in top tier journals, the mobile payment adoption have been
investigated with ungenerous models such as studies of Leong et al. [4], Slade et al. [8] and Tan et al. [9].
Therefore, an holistic research that adapts universal technology acceptance model like UTAUT2 of Venkatesh et
* Corresponding author. E-mail address: nhan.trandanh@gmail.com
123
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
al. [10] is needed. Furthermore, in researching toward mobile payment adoption, based on the nature of e-
payment, a risk consideration extension for UTAUT2 is also required.
In this studies, we examined all factors that could affect to consumer’s intention toward using mobile
payment as well as their actual usage behaviours based on an integration between the UTAUT2 model and
perceived risk extension. Besides, the impacts of different individual risk facets were also investigated in this
research. The paper is structured as follows. In the next sections we will discuss the conceptual model
development with proposed hypothesis, methodology, results, and conclusion.
2. Conceptual model development
2.1. Performance expectancy
Performance expectancy refers to an individual’s perception that using a technology could provide benefits to
users in performing certain activities [10]. Reflecting a range of attributes that a technology could give benefits
to clients, performance has been conceptualized by using system features that could enhance speed, productivity,
and chances of task accomplishment as well as perceived usefulness [10, 11]. Five constructs including
perceived usefulness that adapted from Technology Acceptance Model - TAM [12], Extended Technology
Acceptance Model - TAM2 [13], and Combined Technology Acceptance Model and Theory of Planned
Behavior – C-TAM-TPB [14]; extrinsic motivation that adapted from Motivation Model – MM [15]; job-fit that
adapted from Model of PC Utilization - MPCU [16]; relative advantage that adapted from Innovation Diffusion
Theory – IDT [17]; and outcome expectations that adapted from Social Cognitive Theory [18, 19].
Unambiguously, in diverse task settings, performance expectancy was affirmed to affect intentions to use
technological systems [20]. Consumers’ perception that utilizing mobile payment would support them to achieve
benefits in executing payment tasks may thus affect the behavioral intention of mobile payment adoption.
Accordingly, the following hypothesis was formulated.
H1. There is a positive relationship between consumers’ performance expectancy of mobile payment and
their intentions to adopt the technology.
2.2. Effort expectancy
Effort expectancy is defined as an individual’s estimation of the effort required to accomplish a task utilizing
a given technology [10]. The concept of effort expectancy has been captured by three remarkably similar
constructs derived from different models comprise Technology Acceptance Model - TAM/TAM2 (perceived
ease of use), Model of PC Utilization (complexity), and Innovation Diffusion Theory (ease of use). Many
researchers such as Venkatesh et al. [11], Venkatesh et al. [10], and Dwivedi et al. [21] have provided
significant evidence to validate the influence of effort expectancy on individual’s intentions to adopt
technological systems. Nevertheless, the relationship between effort expectancy and intentions has become
inconclusive due to the recent research that found non-significant relationships or relationships with low
magnitude between perceived effortlessness and behavioral intentions [20, 22, 23]. In the context of mobile
payment, most of modern applications have been designed to become effortless payment methods with intuitive
tools and easy-to-follow operating procedures toward the general population; therefore, consumers’ competence
to operate such mobile payment systems could be considered as a primeval antecedent to develop behavioral
intentions. In view of that, the following hypothesis was developed.
H2. There is a positive relationship between consumers’ effort expectancy of mobile payment and their
intentions to adopt the technology.
2.3. Social influences
Social influence is defined as the degree to which an individual perceives that essential referents such as
family and friends believe he or she should adopt a specific technology [10]. Three constructs related to social
influence are subjective norm (derived from Theory of Reasoned Action - TRA of , Extended Technology
Acceptance Model – TAM2 of Venkatesh and Davis [13], Theory of Planned Behavior TPB/DTPB of Ajzen
[24], and Combined Technology Acceptance Model and Theory of Planned Behavior – C-TAM-TPB of Taylor
and Todd [14]), social factors (derived from Model of PC Utilization – MPCU of Thompson et al. [16]), and
image (derived from Innovation Diffusion Theory – IDT of Moore and Benbasat [17]). Even though they have
diverse names, each of the aforementioned constructs contains the explicit or implicit concepts of the impact on
individual’s adopting intention by the manner in which they consider others’ thinking of them because of
utilizing a particular technology. That impact has been validated by many studies such as Venkatesh et al. [11]
124
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
and Venkatesh et al. [10]. Furthermore, the current trend of social commerce that intensively utilizing
individuals’ social connection to promote innovative technology such as mobile payment will make that
relationship become more relevant to consumers. Based on the discussion above, the following hypothesis was
developed.
H3. There is a positive relationship between consumers’ social influences regarding mobile payment and their
intentions to adopt the technology.
2.4. Facilitating conditions
In business-to-consumer (B2C) settings, facilitating conditions represent the degree to which an individual
believes that supporting resources and people are available to facilitate task accomplishment based on a
particular utilized technology [10]. This notion incorporated three different constructs derived from various
models that operationalized to embrace traits of the technical and/or managerial settings considered to eliminate
obstacles of technological usage. Those three constructs are: perceived behavioral control that adapted from
Theory of Planned Behavior TPB/DTPB [24] and Combined Technology Acceptance Model and Theory of
Planned Behavior – C-TAM-TPB [14]; facilitating conditions that adapted from Model of PC Utilization –
MPCU [16]; and compatibility that adapted from Innovation Diffusion Theory – IDT [17]. Venkatesh et al. [10]
suggested that an individual provided with an advantageous set of facilitating conditions is more likely to have a
greater intention to utilize a specific technology. Their prior research also found that facilitating conditions have
significant direct impact on usage behavior toward the technology [11]. Regarding to the most essential resource
that an individual need to obtain to perform mobile payment tasks, smart mobile devices (e.g. smartphones,
tables, smart watches, etc.), they currently are becoming more and more popular and inexpensive for consumers
to afford. Additionally, supporting knowledge and people are getting easier to interact based on the current
revolution of Industry 4.0. The aforementioned developments could boost both consumers’ behavioral intention
and usage behavior toward mobile payment. Therefore, the following hypothesis was formulated.
H4a. There is a positive relationship between consumers’ facilitating conditions regarding mobile payment
and their intentions to adopt the technology.
H4b. There is a positive relationship between consumers’ facilitating conditions regarding mobile payment
and their usage of the technology.
2.5. Hedonic Motivation
Hedonic motivation is defined as the extent to which individual believe that utilizing a technology could
provide fun or pleasure [10]. This construct has been developed based on the extended finding of technological
adoption theories that consumers would not only utilize a specific technology to accomplish tasks but also to
entertain. Accordingly, adoption literature has been extended from merely focusing on internal beliefs and
utilitarian factors to incorporating playfulness, entertainment value, and enjoyment [21, 25, 26]. In the consumer
context of technology adoption research, hedonic motivation has also been affirmed as a significant factor of
technological behavioral intentions [10, 27, 28]. Akin to most recent B2C applications, mobile payment
applications are designed to comprise hedonic features by interactive visualized interfaces or gamification.
Those features could provide consumers more incentives to use mobile payment as well as enhance their
engagement with the technology. Thus, the following hypothesis was developed.
H5. There is a positive relationship between consumers’ hedonic motivation of mobile payment and their
intentions to adopt the technology.
2.6. Price Value
Venkatesh et al. [10] define price value as the degree to which an individual believes that using a technology
could make him or her face a cognitive trade-off between perceived benefits and monetary cost of using the
technology. This construct has been developed by adapting the marketing literature which conceptualize
monetary cost/price in conjunction with the quality of products or services to ascertain their perceived value [29,
30]. In mobile payment context, monetary cost could include elements such as data service carriers costs (mobile
Internet), mobile device cost, annual or monthly subscription fees, and transaction fees, where applicable.
Consumers are more likely to adopt the technology if they perceived that benefits obtained could overcome
associated monetary cost from usage, in other words the price value is positive. Consequently, price value could
be formulated as a predictor of behavioral intention to utilize mobile payment and the following hypothesis was
developed.
125
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
H6. There is a positive relationship between consumers’ price value of mobile payment and their intentions to
adopt the technology.
2.7. Habit
Habit is defined as the degree to which an individual tend to perform behaviors automatically as a result of
learning from prior experiences [10, 31]. Besides, Venkatesh et al. [10] have distinguished habit from experience
with two fundamental dissimilarities. The first dissimilarities is that to form habit, experience is a compulsory
condition but not sufficient. The second dissimilarities is that divergent levels of habit could be formed as a
result of similar experience (regarded as passage of chronological time) along with different degree of interaction
and familiarity toward the technology. Researchers also suggested that habit should be operationalized as a self-
reported perception [10, 31]. The effects of habit on both behavioral intention and actual usage behavior have
been proposed and validated by researchers such as Pavlou and Fygenson [32], Lankton et al. [33], and
Venkatesh et al. [10]. With the current development of mobile commerce, consumers are using mobile devices
like smartphones for almost daily tasks such as internet surfing, social connecting, text and voice chatting, video
calling, gaming, shopping, bill paying, etc. Therefore, that development could make mobile-related habits
relatively more relevant than before toward the influence on both behavioral intention and actual usage behavior
of mobile payment. Accordingly, the following proposition was developed.
H7a. There is a positive relationship between consumers’ habit of mobile payment and their intentions to
adopt the technology.
H7b. There is a positive relationship between consumers’ habit of mobile payment and their usage of the
technology.
2.8. Perceived Risk
According to Featherman and Pavlou [34], perceived risk involves various individual risk aspects that
encompass performance risk, financial risk, time risk, psychological risk, social risk, and privacy risk. The
researchers also examine construct overall risk as a composite of the individual risk aspects. Performance risk is
defined as the likelihood that the technological system being interrupted or not operating in the way it was
designed and publicized to be, and thus inadequately provide the preferred outcome [34]. In mobile payment
context, consumers could face with the undesired probability that the system not performing well and creating
problems with their credit or the built-in security features could not be strong enough to protect their account or
the mobile payment servers not performing well and process payments incorrectly. Financial risk reflects the
prospective monetary expenditure accompanied by the consumption and maintenance of a product or service as
well as the recurrent possibility of financial loss as a result of fraud [34]. Regarding to mobile payment,
consumers could face monetary loss due to their improper operating, system inaccuracy or potential fraud. As a
results, their non-refundable paying money would be sent to wrong receivers or scammers or just be disappeared.
Time risk is defined as the potential time loss of researching, operating purchasing process, learning how to
utilize a product or service that an individual may suffer if he or she has to replace it when its performance does
not meet with prior expectations [34]. When adopting mobile payment, consumer would spend time setup the
new application, to learning how to operate that application and may spend more time for fixing payment errors
when using. The time investment could be too high for consumers to adopt a new e-service like mobile payment.
Furthermore, if they find that mobile payment is not a proper payment method for them that require switching to
another payment method, they will lose all their prior time investment. Psychological risk reflects the possibility
that process of selecting the producer or operating with product or service will negatively affect consumer’s
peacefulness and self-perception or the possibility that frustrating of not attaining a performance goal could
make consumer suffer a loss of self-esteem or ego [34]. In mobile payment context, consumer may suffer
uncomfortable feeling, anxiety or tension when signing up for and using the technology. Social risk reflects the
possibility of losing consumer’s status in a social group due to utilizing a product or service [34]. Consumers
may concern about the situation that others would think less highly of them if they make mistakes when using
mobile payment. Privacy risk is defined as the possibility that consumer could losing control over their personal
information in case of those private information could be utilized without their knowledge or permission, or in
more extreme case, their identity information could be used by criminals to conduct fraudulent dealings [34]. By
using mobile payment, consumers could be suffered the risk of losing their payment information that disclosed to
unwanted person or losing control of their checking account due to Internet hacking. Finally, overall risk is a
universal measurement of perceived risk when all individual risk aspects are assessed together [34]. Based on the
aforementioned discussion, we can hypothesize that:
126
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
H8a. There is a positive relationship between consumers’ perceived performance risk toward mobile payment
and their overall perceived risk of adopting the technology.
H8b. There is a positive relationship between consumers’ perceived financial risk toward mobile payment and
their overall perceived risk of adopting the technology.
H8c. There is a positive relationship between consumers’ perceived time risk toward mobile payment and
their overall perceived risk of adopting the technology.
H8d. There is a positive relationship between consumers’ perceived psychological risk toward mobile
payment and their overall perceived risk of adopting the technology.
H8e. There is a positive relationship between consumers’ perceived social risk toward mobile payment and
their overall perceived risk of adopting the technology.
H8f. There is a positive relationship between consumers’ perceived privacy risk toward mobile payment and
their overall perceived risk of adopting the technology.
Featherman and Pavlou [34] also posited that individual’s perceived risk of using a technology has negative
influence on perceived usefulness (equivalent to performance expectancy) and behavioral intention toward that
technology while his or her perceived ease of use (equivalent to effort expectancy) also has adverse impact on
the perceived risk. In mobile payment context, it is likely that the lower risk consumers notice from utilizing
mobile payment, the higher tendency they have to perceive mobile payment as beneficial and adoptable. In
addition, it is also expected that with lower perceive effort consumers would have higher tendency to perceive
mobile payment as a risk-free technology. Consequently, the following hypotheses were developed.
H9a. There is a negative mediating effect of consumers’ overall perceived risk of adopting mobile payment
on the relationship between their effort expectancy and intentions to use the technology.
H9b. There is a negative mediating effect of consumers’ performance expectancy of mobile payment on the
relationship between their overall perceived risk and intentions to use the technology.
H9c. There is a negative relationship between consumers’ overall perceived risk of adopting mobile payment
and their intentions to use the technology.
2.9. Behavioral Intention
According to Davis [35], behavioral intention is defined as the degree to which an individual believes that
they will implement a particular behavior. In technology adoption theories, the relationship between behavioral
intention and usage behavior has been consistently confirmed (e.g., Davis et al. [36], Venkatesh and Davis [13],
Venkatesh et al. [11], Venkatesh et al. [10], Baptista and Oliveira [20]). In mobile payment context, there is still
a gap between intention and actual use toward the technology. Although the development of mobile commerce
has led consumers to use mobile devices in various aspects of their daily lives, the actual amount of mobile
payment is still small compared to other forms of payment. In accordance with the technological adoption
literature, it can be hypothesized that:
H10. There is a positive relationship between consumers’ intentions to adopt mobile payment and their actual
usage of the technology.
All in all, based on the aforementioned discussion, the conceptual model has been developed and shown in
Fig. 1
127
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
Performance
Risk
Financial
Risk
Time
Risk
Psychological
Risk
Social
Risk
Privacy
Risk
H8a
H8b
H8c
H8d
H8e
H8f
H9b
Performance
Expectancy
Perceived
Risk
H1
H9a
H9c
Effort
Expectancy
H2
Behavioral
Intentions
H10
Usage
Behavior
H3
Social
Influence
H4a
H4b
Facilitating
Conditions
H5
Hedonic
H6
Motivation
H7a
H7b
Price
Value
Habit
Fig. 1. Conceptual Model.
3. Methods
3.1. Measurement instruments
A questionnaire-based survey was developed in order to test the theoretical constructs. Constructs and
measurement items were adapted with slight modifications from technology acceptance literature to build the
questionnaire. Measurement items for constructs of performance expectancy, effort expectancy, social influence,
facilitating conditions, hedonic motivation, price value, habit, and behavioral intention are adapted from
Venkatesh et al. [10]. Constructs of perceived risk and its individual risk facets were operationalized by items
adapted from Featherman and Pavlou [34]. Finally, Im et al. [37] was adapted to operationalize the usage
behaviour construct. Except usage behaviour construct, all main measurement items were measured on a five-
point Likert scale, ranging from totally disagree (1) to totally agree (5). The usage behaviour measurement was
operationalized by one item that measure consumers’ actual frequencies of mobile payment usage (have not
used, once a year, once in six months, once in three months, once a month, once a week, once in 4–5 days, once
in 2–3 days, and almost every day). The items for all constructs are presented in the Appendix A. Two
demographic variables related to age and gender were also included in the questionnaire. Age was measured in
years and gender was coded using a 0 or 1 dummy variable where 1 represented women.
The questionnaire was primarily developed in English, based on the literature with reviewing for content
validity experts from a university. Because the data collection procedure was operated in Vietnamese context,
then later all English instruments was translated into Vietnamese language by a professional translator. The
questionnaire was built online with Google Form service.
3.2. Data collection
Eight hundred and forty-three (843) students and alumni from universities in Vietnam were contacted by e-
mail and social network account in June of 2018. A hyperlink to the online survey was included in the messages.
Three hundred and twenty-nine (329) valid responses were received. The overall response rate was 39%, which
is reasonable for studies of this scale. 81% of the subjects were females. Because of our convenience sampling,
this gender distribution in the sample could be results of that fact that women are have more interest on mobile
128
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
shopping and mobile payment than men and more willing to answer the questionnaire. The age ranged from 18
to 26 years. Individuals which are university students accounted for 82.4% of the data. The sample is an
indicative group to test the instrument because university students have high potential to adopt new mobile
technologies such as mobile payment [38]. There are 10% of the subjects have not used mobile payment and the
highest portion of mobile usage is once a month.
4. Results
4.1. Measurement model
In order to evaluate the constructs’ reliability, Cronbach’s Alpha reliability test was utilized. It is a widely
used measure that examines the scale reliability, so called the internal consistency analysis [39]. By measuring
the reliability coefficient, the reliability test could assess the consistency of the entire scale. According to
Nunnally [40] a scale would be high reliable level if the coefficient alpha is greater than 0.7 while the coefficient
alpha is higher than 0.6 means the scale is reliable. As seen from Table 1, Cronbach’s Alpha of all constructs are
above the expected threshold of 0.7, showing evidence of internal consistency. These constructs would be
utilized in further analysis to test the proposed hypotheses. The means and standard deviations of each constructs
as well as their belonging measurement items are also shown in Table 1.
129
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
Table 1. Means, standard deviations, and constructs’ reliability for the measurement model.
Corrected Item-Total
Mean
SD
Cronbach's Alpha
Constructs and Items
Correlation
3.97
3.95
4.14
3.84
3.93
3.82
3.82
3.82
3.85
3.80
3.49
3.49
3.43
3.55
3.80
3.71
3.81
3.95
3.71
3.07
2.97
3.47
2.78
3.54
3.47
3.54
3.63
2.67
2.64
2.32
3.07
2.66
2.23
2.35
2.37
1.96
2.28
2.17
2.84
2.88
2.85
3.13
2.53
0.76
0.88
0.83
0.88
0.88
0.81
0.90
0.86
0.94
0.96
0.80
0.95
0.93
0.92
0.70
0.98
0.89
0.92
0.94
0.83
1.00
0.94
0.96
0.68
0.79
0.77
0.79
0.91
1.13
0.97
1.13
1.12
0.75
0.89
0.91
0.89
0.92
0.90
0.74
1.03
1.05
0.99
0.93
0.895
Performance Expectancy (PE)
0.721
0.802
0.792
0.755
PE01
PE02
PE03
PE04
0.906
Effort Expectancy (EE)
0.770
0.755
0.800
0.831
EE01
EE02
EE03
EE04
Social Influence (SI)
0.823
0.736
0.708
0.695
0.634
SI01
SI02
SI03
Facilitating Conditions (FC)
0.577
0.587
0.617
0.348
FC01
FC02
FC03
FC04
0.816
0.835
0.852
Hedonic Motivation (HM)
0.736
0.608
0.663
HM01
HM02
HM03
Price Value (PV)
PV01
0.682
0.742
0.666
PV02
PV03
Habit (HT)
HT01
0.770
0.725
0.520
0.782
HT02
HT03
HT04
Perceived Risk (PR)
PR01
0.888
0.780
0.733
0.750
0.734
0.675
0.750
PR02
PR03
PR04
PR05
Performance risk (PER)
PER01
0.586
0.572
0.506
0.568
PER02
PER03
PER04
130
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
2.79
2.72
2.56
2.98
2.56
2.78
2.31
2.53
2.55
2.44
2.52
1.93
1.91
1.95
1.72
1.67
1.94
2.35
2.09
1.78
1.65
1.55
2.14
3.36
3.22
3.40
3.46
3.78
4.05
3.73
3.57
4.08
1.03
0.84
0.96
1.04
1.07
1.01
0.84
1.23
1.13
1.13
1.13
0.94
1.01
0.76
0.90
0.88
1.06
1.05
1.06
0.78
0.90
0.82
1.15
0.96
1.08
1.17
1.05
0.74
0.85
0.91
0.92
1.89
0.542
PER05
Financial risk (FIR)
FIR01
0.846
0.861
0.621
0.693
0.676
0.742
FIR02
FIR03
FIR04
Time risk (TIR)
TIR01
0.670
0.590
0.710
0.741
0.608
0.608
TIR02
TIR03
TIR04
TIR05
TIR06
Psychological risk (PSR)
PSR01
0.816
0.584
0.575
0.618
0.657
0.607
PSR02
PSR03
PSR04
PSR05
Social risk (SOR)
SOR01
0.739
0.834
0.773
0.660
0.637
0.451
SOR02
SOR03
Privacy risk (PRR)
PRR01
0.724
0.729
0.639
PRR02
PRR03
Behavioral Intentions (BI)
BI01
0.561
0.623
0.644
BI02
BI03
Usage Behavior (UB)
4.2. Hypotheses testing
To test hypotheses of the relationships between the overall perceived risk and its individual risk aspects, the
Linear regression analysis was utilized. In the test, the overall perceived risk construct was treated as a
dependent variable; the mean score of the overall perceived risk construct was regressed across mean scores for
six independent variables of individual risk constructs. The regression results shown in Table 2 indicated that
there are statistically significant positive impacts of consumers’ perceived psychological risk (β = 0.277, p <
0.001), time risk (β = 0.261, p < 0.001), performance risk (β = 0.191, p = 0.001) and privacy risk (β = 0.088, p =
0.021) toward mobile payment on their overall perceived risk of adopting the technology. In the other hand,
based on the regression results, influences of consumers’ perceived financial risk and social risk are not
statistically significant. Therefore, the hypotheses H8a, H8c, H8d and H8f are supported while the hypotheses
H8b and H8e are not supported.
Table 2. Regression analysis results for testing relationships between individual risks and perceived risk.
Regression Analysis
Independent Variable
Coefficients
t-value
p-value
F-value
p-value R-square
131
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
(Constant)
-0.028
0.191
0.057
0.261
0.277
0.064
0.088
-0.206
3.423
1.239
6.078
5.303
1.420
2.325
0.837
0.001
Performance risk (PER)
Financial risk (FIR)
Time risk (TIR)
0.216
< 0.001
< 0.001
0.156
Psychological risk (PSR)
Social risk (SOR)
Privacy risk (PRR)
Overall model
0.021
60.958
< 0.001
0.532
In order to test the proposed mediating effects, Baron and Kenny [41]’s three-step mediation testing approach
was adapted with a bootstrapping integrated enhancement provided by Preacher and Hayes [42]. The modern
method could be used to replace the traditional mediation testing method proposed by Sobel [43] by overcome
the assumption that the product of coefficients constituting the indirect effect must be normally distributed. In
behavioral studies, that assumption is usually violated due to the fact that the distribution tends to be skewed and
leptokurtic [42]. According to the researchers, a statistically significant simple mediation effect could be shown
if the values between the upper and lower limits of confidence interval for the size of the indirect path do not
include zero.
As show in Table 3, the Effort Expectancy negatively significantly predicts the Perceived Risk ( = -0.242, t
= -4.896, p < 0.001). Additionally, when the Perceived Risk is not in the model, the Effort Expectancy
significantly predicts Behavioral Intentions ( = 0.390, t = 8.516, p < 0.001). Besides, the results also indicate
that Effort Expectancy significantly predicts Behavioral Intentions even with Perceived Risk in the model ( =
0.360, t = 0.047, p < 0.001); Perceived Risk also significantly predicts Behavioral Intentions ( = -0.123, t = -
2.409, p = 0.017). The Effort Expectancy predict Behavioral Intentions less strongly in the model of step 3 than
in the model of step 2 ( = 0.360 < = 0.390). These results satisfied the partial mediation conditions suggested
by Baron and Kenny [41]. Moreover, the indirect effect of Effort Expectancy on Behavioral Intentions was
estimated with = 0.030 and bias-corrected bootstrapped confidence interval = [0.006, 0.064] (BCa
bootstrapped CI based on 1000 samples). This range does not include zero suggest that there is likely to be a
genuine indirect effect, or in other words, Perceived Risk is a mediator of the relationship between Effort
Expectancy and Behavioral Intentions. Therefore, the hypothesis H8a is supported.
132
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
Table 3. Mediating effect of Perceived Risk (M) on Effort Expectancy (X) and Behavioral Intentions (Y)
Step 1: Regression predicting M from X (R2 = 0.068, p < 0.001)
Coefficients
3.152
SE
0.193
0.049
t-value
16.320
-4.896
p-value
<0.001
<0.001
(Constant)
Effort Expectancy (EE)
-0.242
Step 2: Regression predicting Y from X (R2 = 0.182, p < 0.001)
Coefficients
SE
0.179
0.046
t-value
12.811
8.516
p-value
<0.001
<0.001
(Constant)
2.292
0.390
Effort Expectancy (EE)
Step 3: Regression predicting Y from X and M (R2 = 0.196, p < 0.001)
Coefficients
2.679
SE
0.239
0.051
0.047
t-value
11.194
-2.409
7.651
p-value
<0.001
0.017
(Constant)
Perceived Risk (PR)
-0.123
Effort Expectancy (EE)
0.360
<0.001
Direct effect of X on Y
SE
Effect
t-value
p-value
7.651
<0.001
Effort Expectancy (EE)
Indirect effect of X on Y
0.360
0.047
Effect
Boot SE
Boot LLCI
Boot ULCI
Effort Expectancy (EE)
0.030
0.014
0.006
0.064
Note: - X: Effort Expectancy (EE); M: Perceived Risk (PR); Y: Behavioral Intentions (BI).
- Number of bootstrap samples for bias corrected bootstrap confidence intervals: 1000
- Level of confidence for all confidence intervals in output: 95.00
The testing results shown in Table 4 indicate that the Perceived Risk negatively significantly predicts the
Performance Expectancy ( = -0.284, t = -5.305, p < 0.001). Furthermore, when the Performance Expectancy is
not in the model, the Perceived Risk negatively significantly predicts Behavioral Intentions ( = -0.224, t = -
4.212, p < 0.001). Besides, the results also indicate that Perceived Risk does not significantly predicts Behavioral
Intentions with Performance Expectancy in the model ( = -0.063, t = -1.379, p = 0.169) while Performance
Expectancy significantly predicts Behavioral Intentions ( = 0.569, t = 12.565, p < 0.001). These results satisfied
the complete mediation conditions suggested by Baron and Kenny [41]. Moreover, the indirect effect of Effort
Expectancy on Behavioral Intentions was estimated with = -0.161 and bias-corrected bootstrapped confidence
interval = [-0.243, -0.086] (BCa bootstrapped CI based on 1000 samples). This range does not include zero
suggest that there is likely to be a genuine indirect effect, or in other words, Effort Expectancy is a mediator of
the relationship between Perceived Risk and Behavioral Intentions. Therefore, the hypothesis H8b is supported.
Table 4. Mediating effect of Performance Expectancy (M) on Perceived Risk (X) and Behavioral Intentions (Y)
Step 1: Regression predicting M from X (R2 = 0.079, p < 0.001)
Coefficients
4.597
SE
0.126
0.053
t-value
36.598
-5.305
p-value
<0.001
<0.001
(Constant)
Perceived Risk (PR)
-0.284
Step 2: Regression predicting Y from X (R2 = 0.051, p < 0.001)
133
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
Coefficients
4.283
SE
0.125
0.053
t-value
34.235
-4.212
p-value
<0.001
<0.001
(Constant)
Perceived Risk (PR)
-0.224
Step 3: Regression predicting Y from X and M (R2 = 0.361, p < 0.001)
Coefficients
1.667
SE
0.232
0.045
0.046
t-value
7.183
p-value
<0.001
<0.001
0.169
(Constant)
Performance Expectancy (PE)
Perceived Risk (PR)
0.569
12.565
-1.379
-0.063
Direct effect of X on Y
SE
Effect
t-value
p-value
-0.063
0.046
-1.379
0.169
Perceived Risk (PR)
Indirect effect of X on Y
Effect
Boot SE
Boot LLCI
Boot ULCI
Perceived Risk (PR)
-0.161
0.039
-0.243
-0.086
Note: - X: Perceived Risk (PR); M: Performance Expectancy (PE); Y: Behavioral Intentions (BI).
- Number of bootstrap samples for bias corrected bootstrap confidence intervals: 1000
- Level of confidence for all confidence intervals in output: 95.00
Regarding to the relationships between Behavioral Intentions toward mobile payment adoption and its
potential predictors, a similar Linear regression analysis procedure was conducted. The regression results shown
in Table 5 indicated that there are statistically significant positive impacts of consumers’ performance
expectancy (β = 0.308, p < 0.001), facilitating conditions (β = 0.186, p = 0.001), price value (β = 0.107, p =
0.041) and habit (β = 0.088, p < 0.001) toward mobile payment on their behavioral intentions of adopting the
technology. In the other hand, based on the regression results, influences of consumers’ effort expectancy, social
influence, hedonic motivation and perceived risk are not statistically significant. Therefore, the hypotheses H1,
H4a, H6 and H7a are supported while the hypotheses H2, H3, H5 and H9c are not supported.
134
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
Table 5. Regression results for testing relationships between behavioral intentions and its potential predictors.
Regression Analysis
Independent Variable
Coefficients
t-value
2.413
6.130
0.643
1.071
3.334
-0.835
2.055
6.139
0.192
p-value
F-value
p-value R-square
(Constant)
0.633
0.016
Performance Expectancy (PE)
Effort Expectancy (EE)
Social Influence (SI)
Facilitating Conditions (FC)
Hedonic Motivation (HM)
Price Value (PV)
0.308
< 0.001
0.521
0.029
0.046
0.285
0.186
0.001
-0.035
0.107
0.405
0.041
Habit (HT)
0.249
< 0.001
0.848
Perceived Risk (PR)
Overall model
0.008
42.805
< 0.001
0.517
A final regression analysis was also conducted to test the relationships between potential influences and usage
behavior of mobile payment, and its results are shown in Table 6. The regression results suggested that usage
behavior are statistically significantly positively influenced by behavioral intentions (β = 0.353, p = 0.017),
facilitating conditions (β = 0.410, p = 0.005), and habit (β = 0.880, p < 0.001).
Table 6. Regression results for testing relationships between usage behavior and its potential predictors.
Regression Analysis
Independent Variable
Coefficients
t-value
-2.272
2.396
2.854
7.668
p-value
F-value
p-value R-square
(Constant)
-1.167
0.024
Behavioral Intentions (BI)
Facilitating Conditions (FC)
Habit (HT)
0.353
0.017
0.410
0.005
0.880
< 0.001
Overall model
62.884
< 0.001
0.367
Fig. 2 illustrates results from all above hypotheses testing.
135
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
Performance
Risk
Financial
Risk
Time
Risk
Psychological
Risk
Social
Risk
Privacy
Risk
H8a
H8b
H8c
H8d
H8e
H8f
H9b
Performance
Expectancy
Perceived
Risk
H1
H9a
H9c
Effort
Expectancy
H2
Behavioral
Intentions
H10
Usage
Behavior
H3
Social
Influence
H4a
H4b
Facilitating
Conditions
H5
Hedonic
H6
Motivation
H7a
H7b
Price
Value
Habit
Fig. 2. Model testing results
5. Limitations and future research
Beside of our study’s main contribution that adds into the existing body of knowledge, we also recognize its
limitations, mostly regarding the sampling with typically young, highly educated people as responders. The
respondents’ behavioral patterns might diverge to some extent in comparison with the population average. With
the behaviors that are mostly more pioneering and rapider to adopt new technologies, this sampling may have
biased the effects. It is likely that seniors and less educated consumers or those who hold low computing or
Internet-related capability would recognize more difficulty in adopting mobile payment and greater intrinsic
mobile payment usage risks. Future research can be constructed based on this study by examining the proposed
model in different age groups or applying this model to other countries and also other contexts.
6. Conclusions
Mobile payment is getting emergent consideration worldwide, however, the impacts of various factors on its
adoption have not yet been expansively investigated. Toward fulfilling this research gap, we had extended the
well-known theories, namely UTAUT2, with perceived risks as well as explored impacts of various individual
risk aspects on the overall perceived risk. Our results confirmed that performance expectancy, facilitating
conditions, price value and habit altogether have direct effects on the intention to adopt mobile payment. To
determine usage behavior of mobile payment, the behavior intention along with facilitating conditions and habit
make impacts on the actual use. Regarding to perceived risk, it does have negative indirect, not direct, impact on
behaviors intention through the mediation of performance expectancy. When perceived risk is included in the
model, the effect of effort expectancy on behavior intention would be diminished due to the mediation role of
perceived risk between the two constructs. Furthermore, in case of mobile payment, it is suggested that the
overall perceived risk would be significantly influenced by perceived psychological risk, time risk, performance
risk, and privacy risk.
Appendix A. Main model measurements
136
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
Constructs
Performance
Expectancy
Items
PE01 -
PE02 -
Source
I find mobile payment useful in my daily life.
Using mobile payment helps me accomplish things more
quickly.
PE03 -
PE04 -
Using mobile payment increases my productivity.
Using mobile payment increases my chances of job done.
Effort
Expectancy
EE01 -
EE02 -
Learning how to use mobile payment is easy for me.
My interaction with mobile payment is clear and
understandable.
EE03 -
EE04 -
I find mobile payment easy to use.
It is easy for me to become skilful at using mobile payment
Social
Influence
SI01 -
SI02 -
SI03 -
People who are important to me think that I should use mobile
payment
People who influence my behavior think that I should use
mobile payment
People whose opinions that I value prefer that I use mobile
payment
Facilitating
Conditions
FC01 -
FC02 -
FC03 -
FC04 -
I have the resources necessary to use mobile payment
I have the knowledge necessary to use mobile payment
Mobile payment is compatible with other technologies I use.
I can get help from others when I have difficulties using
mobile payment
Hedonic
HM01 - Using mobile payment is fun.
Motivation
HM02 - Using mobile payment is enjoyable.
HM03 - Using mobile payment is very entertaining.
Price Value
Habit
PV01 -
PV02 -
PV03 -
Mobile payment is reasonably priced.
Mobile payment is a good value for the money.
At the current price, mobile payment provides a good value.
HT01 - The use of mobile payment has become a habit for me.
HT02 - I am addicted to using mobile payment.
HT03 - I must use mobile payment.
HT04 - Using mobile payment has become natural to me.
Performance
Risk
PER01 - The mobile payment might not perform well and create
problems with my credit.
PER02 - The security systems built into the mobile payment system are
not strong enough to protect my checking account.
PER03 - The probability that there will be something wrong with the
performance of the mobile payment or that it will not work
properly is high
PER04 - Considering the expected level of service performance of the
mobile payment, for me to sign up for and use it would be
risky
PER05 - Mobile payment servers may not perform well and process
payments incorrectly.
Financial
Risk
FIR01 - The chances of losing money if I use mobile payment are high
FIR02 - Using a mobile payment service subjects my checking account
to potential fraud.
FIR03 - My signing up for and using a mobile payment would lead to
a financial loss for me.
FIR04 - Using a mobile payment service subjects your checking
account to financial risk.
Time Risk
TIR01 - If I use a new mobile payment method, the chances that I will
lose time due to having to switch to a different payment
137
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
method is high
TIR02 - My signing up for and using mobile payment would lead to a
loss of convenience of me because I would have to waste lot
of time fixing payments errors.
TIR03 - The investment of my time involved to switch to (and set up)
a new mobile payment method is high.
TIR04 - The possible time loss from having to set-up and learn how to
use a new mobile payment method is high.
TIR05 - Using mobile payment could lead to an inefficient use of my
time.
TIR06 - Using mobile payment could require more time than when not
using them.
Psychological PSR01 - Mobile payment will not fit in well with my self-image or self-
Risk
concept.
PSR02 - The usage of mobile payment would lead to a psychological
loss for me because it would not fit in well with my self-image
or self-concept.
PSR03 - My signing up for and using mobile payment makes me feel
uncomfortable.
PSR04 - My signing up for and using mobile payment gives me an
unwanted feeling of anxiety.
PSR05 - My signing up for and using mobile payment causes me to
experience unnecessary tension.
Social Risk
Privacy Risk
Overall Risk
SOR01 - The chances that signing up for and using mobile payment
will negatively affect the way others think of me is high
SOR02 - My signing up for and using mobile payment would lead to a
social loss for me because my friends and relatives would
think less highly of me.
SOR03 - I would be concerned about what others would think of me if I
made a bad choice when using mobile payment.
PRR01 - The chances that using mobile payment will cause me to lose
control over the privacy of your payment information is high
PRR02 - My signing up for and using mobile payment would lead to a
loss of privacy for me because my personal information would
be used without my knowledge.
PRR03 - Internet hackers (criminals) might take control of my checking
account if I use mobile payment
OR01 - On the whole, considering all sorts of factors combined, I
would say it would be risky to sign up for and use mobile
payment
OR02 - Using mobile payment to finish my transactions would be
risky.
OR03 - Mobile payment are dangerous to use.
OR04 - Using mobile payment would add great uncertainty to my
payment.
OR05 - Using mobile payment exposes me to an overall risk.
Behavior
Intentions
BI01 -
BI02 -
BI03 -
I intend to continue using mobile payment in the future.
I will always try to use mobile payment in my daily life.
I plan to continue to use mobile payment frequently.
Usage
Behavior
UB
What is your actual frequency of use of Internet banking
services?
138
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
References
[1] Antonio, G., et al., (2010). Mobile payment applications: offer state of the art in the Italian market. info,
12(5), 3-22.
[2] Pham, T.-T.T. and J.C. Ho, (2015). The effects of product-related, personal-related factors and
attractiveness of alternatives on consumer adoption of NFC-based mobile payments. Technology in Society,
43, 159-172.
[3] Aik-Chuan, T., et al., (2015). The effects of convenience and speed in m-payment. Industrial Management &
Data Systems, 115(2), 311-331.
[4] Leong, L.-Y., et al., (2013). Predicting the determinants of the NFC-enabled mobile credit card acceptance:
A neural networks approach. Expert Systems with Applications, 40(14), 5604-5620.
[5] Duane, A., P. O'Reilly, and P. Andreev, (2011). Trusting M-Payments - Realising the Potential of Smart
Phones for M-Commerce: A Conceptual Model & Survey of Consumers in Ireland. Vol. 2.
[6] Merritt, C., (2011). Mobile money transfer services: The next phase in the evolution of person-to-person
payments. Journal of Payments Strategy & Systems, 5(2), 143-160.
[7] Statista. (2018). Total revenue of global mobile payment market from 2015 to 2019. Available from:
https://www.statista.com/statistics/226530/mobile-payment-transaction-volume-forecast/. Accesed 2018.06.30
[8] Slade, E., et al., (2015). Exploring consumer adoption of proximity mobile payments. Journal of Strategic
Marketing, 23(3), 209-223.
[9] Tan, G.W.-H., et al., (2014). NFC mobile credit card: The next frontier of mobile payment? Telematics and
Informatics, 31(2), 292-307.
[10]Venkatesh, V., J.Y.L. Thong, and X. Xu, (2012). Consumer acceptance and use of information technology:
extending the unified theory of acceptance and use of technology. MIS Q., 36(1), 157-178.
[11]Venkatesh, V., et al., (2003). User Acceptance of Information Technology: Toward a Unified View. MIS
Quarterly, 27(3), 425-478.
[12]Davis, F.D., (1989). Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information
Technology. MIS Quarterly, 13(3), 319-340.
[13]Venkatesh, V. and F.D. Davis, (2000). A Theoretical Extension of the Technology Acceptance Model: Four
Longitudinal Field Studies. Management Science, 46(2), 186-204.
[14]Taylor, S. and P. Todd, (1995). Assessing IT Usage: The Role of Prior Experience. MIS Quarterly, 19(4),
561-570.
[15]Davis, F.D., R.P. Bagozzi, and P.R. Warshaw, (1992). EXTRINSIC AND INTRINSIC MOTIVATION TO
USE COMPUTERS IN THE WORKPLACE. Journal of Applied Social Psychology, 22(14), 1111-1132.
[16]Thompson, R.L., C.A. Higgins, and J.M. Howell, (1991). Personal computing: toward a conceptual model
of utilization. MIS quarterly, 125-143.
[17]Moore, G.C. and I. Benbasat, (1996). Integrating diffusion of innovations and theory of reasoned action
models to predict utilization of information technology by end-users, in Diffusion and adoption of
information technology. Springer. 132-146.
[18]Compeau, D.R. and C.A. Higgins, (1995). Computer self-efficacy: Development of a measure and initial
test. MIS quarterly, 189-211.
[19]Compeau, D., C.A. Higgins, and S. Huff, (1999). Social cognitive theory and individual reactions to
computing technology: A longitudinal study. MIS quarterly, 145-158.
[20]Baptista, G. and T. Oliveira, (2015). Understanding mobile banking: The unified theory of acceptance and
use of technology combined with cultural moderators. Computers in Human Behavior, 50, 418-430.
[21]Dwivedi, Y.K., et al., (2016). A generalised adoption model for services: A cross-country comparison of
mobile health (m-health). Government Information Quarterly, 33(1), 174-187.
[22]Lian, J.-W., (2015). Critical factors for cloud based e-invoice service adoption in Taiwan: An empirical
study. International Journal of Information Management, 35(1), 98-109.
[23]Pascual-Miguel, F.J., Á.F. Agudo-Peregrina, and J. Chaparro-Peláez, (2015). Influences of gender and
product type on online purchasing. Journal of Business Research, 68(7), 1550-1556.
[24]Ajzen, I., (1991). The theory of planned behavior. Organizational behavior and human decision processes,
50(2), 179-211.
[25]Thong, J.Y., S.-J. Hong, and K.Y. Tam, (2006). The effects of post-adoption beliefs on the expectation-
confirmation model for information technology continuance. International Journal of Human-Computer
Studies, 64(9), 799-810.
[26]Turel, O., A. Serenko, and N. Bontis, (2007). User acceptance of wireless short messaging services:
Deconstructing perceived value. Information & Management, 44(1), 63-73.
[27]Brown, S.A. and V. Venkatesh, (2005). A model of adoption of technology in the household: A baseline
139
Nhan Tran Danh, Ha Tran Thi Phuong/ MICA 2018 Proceedings
model test and extension incorporating household life cycle. Management Information Systems Quarterly,
29(3), 11.
[28]Childers, T.L., et al., (2001). Hedonic and utilitarian motivations for online retail shopping behavior.
Journal of retailing, 77(4), 511-535.
[29]Zeithaml, V.A., (1988). Consumer perceptions of price, quality, and value: a means-end model and
synthesis of evidence. The Journal of marketing, 2-22.
[30]Dodds, W.B., K.B. Monroe, and D. Grewal, (1991). Effects of price, brand, and store information on
buyers' product evaluations. Journal of marketing research, 307-319.
[31]Limayem, M., S.G. Hirt, and C.M. Cheung, (2007). How habit limits the predictive power of intention: The
case of information systems continuance. MIS quarterly, 705-737.
[32]Pavlou, P.A. and M. Fygenson, (2006). Understanding and predicting electronic commerce adoption: An
extension of the theory of planned behavior. MIS quarterly, 115-143.
[33]Lankton, N.K., E.V. Wilson, and E. Mao, (2010). Antecedents and determinants of information technology
habit. Information & Management, 47(5-6), 300-307.
[34]Featherman, M.S. and P.A. Pavlou, (2003). Predicting e-services adoption: a perceived risk facets
perspective. International journal of human-computer studies, 59(4), 451-474.
[35]Davis, F.D., (1985). A technology acceptance model for empirically testing new end-user information
systems: Theory and results. Massachusetts Institute of Technology.
[36]Davis, F.D., R.P. Bagozzi, and P.R. Warshaw, (1989). User acceptance of computer technology: a
comparison of two theoretical models. Management science, 35(8), 982-1003.
[37]Im, I., S. Hong, and M.S. Kang, (2011). An international comparison of technology adoption: Testing the
UTAUT model. Information & management, 48(1), 1-8.
[38]Sohn, S.Y. and Y. Kim, (2008). Searching customer patterns of mobile service using clustering and
quantitative association rule. Expert Systems with Applications, 34(2), 1070-1077.
[39]Hair, J.F., et al., (2014). Multivariate data analysis. Harlow: Pearson Education Limited.
[40]Nunnally, J.C., (1978). Psychometric methods. New York: McGraw-Hill.
[41]Baron, R.M. and D.A. Kenny, (1986). The moderator–mediator variable distinction in social psychological
research: Conceptual, strategic, and statistical considerations. Journal of personality and social
psychology, 51(6), 1173.
[42]Preacher, K.J. and A.F. Hayes, (2008). Asymptotic and resampling strategies for assessing and comparing
indirect effects in multiple mediator models. Behavior research methods, 40(3), 879-891.
[43]Sobel, M.E., (1982). Asymptotic confidence intervals for indirect effects in structural equation models.
Sociological methodology, 13(1982), 290-312.
140
Bạn đang xem tài liệu "Role of perceived risk in determining consumer acceptance of mobile payment: An empirical study in Vietnam", để tải tài liệu gốc về máy hãy click vào nút Download ở trên
File đính kèm:
- role_of_perceived_risk_in_determining_consumer_acceptance_of.pdf