(Gu al., 2014; Zhou, 2011). Therefore, the TAM

(Gu al., 2014; Zhou, 2011). Therefore, the TAM

(Gu et al., 2009; Luarn and Lin, 2005; Riquelme and Rios, 2010; Shareef et al., 2014; Zhou,
2011). Therefore, the TAM has been modified and extended by including other factors
such as perceived risk (PR) (Hanafizadeh et al., 2014; Riquelme and Rios, 2010), trust
(Kim et al., 2009; Zhou, 2011), and SE (Gu et al., 2009; Luarn and Lin, 2005).
Given that PR has been recognised as one of the most important and frequently
found factors influencing customers’ intention and adoption of different electronic
banking channels (Curran and Meuter, 2005; Eriksson et al., 2008; Gan et al., 2006;
Jaruwachirathanakul and Fink, 2005; Kolodinsky et al., 2004; Taylor and Strutton,
2010), PR was included as an extension to the TAM factors in the same conceptual
model. Further justifications supporting and including PR has been provided in
Section 3.3. Furthermore, in order to provide a better picture explaining the Jordanian
customers’ perception and intention towards MB, there was a necessity to include the
role of personality factors among the conceptual model. Indeed, SE has been noticed
over the related literature as one of the most repeated factors impacting the customers’
perception towards such novel technologies (i.e. Luarn and Lin, 2005; Gu et al., 2009;
Püschel et al., 2010; Zhou, 2012). SE therefore was comprised among the conceptual
model along with TAM factors and PR. Further discussions regarding the research
hypotheses are provided in the following subsections.
3.1 PU
PU is conceptualised as “the degree to which a person believes that using a particular
system would enhance his or her job performance” (Davis et al., 1989, p. 320). Over the
prior literature, PU has been noticed as one of the most influential drivers of BI to adopt
MB (Akturan and Tezcan, 2012; Chen et al., 2014; Hanafizadeh et al., 2014; Kapoor et al.,
2014; Luarn and Lin, 2005; Wessels and Drennan, 2010; Williams et al., 2015). Such as,
PU was empirically supported by Wessels and Drennan (2010) as a key factor
predicting BI to adopt MB by Australian customers. Gu et al. (2009) also empirically
supported the role of PU in contributing to the customers’ willingness to use MB.
More recently, Hanafizadeh et al. (2014) supported the crucial role of PU in motivating
Iranian customers to adopt MB.
BI Adoption
Sources: Adapted from Compeau and Higgins (1995), Davis et al. (1989), and
Featherman and Pavlou (2003)
Figure 1.
Proposed conceptual
model of the
adoption of MB by
Jordanian customers
adoption of
MB in Jordan
Downloaded by New York University At 01:34 05 February 2016 (PT)
According to the model of PC utilisation, the actual use behaviour could be directly
influenced by the role perceived consequences (i.e. PU, job-fit) (Thompson et al., 1991).
With reference to Triandis (1977), individual behaviour is usually determined by
potential behavioural outcomes that are attractive and are more likely to be attended.
Al-Qeisi and Abdallah (2013) supported this assumption by confirming a strong
relationship between performance expectancy (similar factor to PU as proposed by
Venkatesh et al. (2003)) and actual usage of internet banking by Jordanian customers.
Likewise, Zhou et al. (2010) empirically approved a significant relationship between
performance expectancy and actual adoption of MB.
Therefore, this study articulates the following hypotheses:
H1. PU will positively influence Jordanian customers’ intention to adopt MB.
H2. PU will positively influence Jordanian customers’ adoption of MB.
3.2 PEOU
Davis et al. (1989, p. 320) defined PEOU as “the degree to which a person believes that
using a particular system would be free of effort”. Due to the particular nature of MB
which requires a certain level of knowledge and skill, PEOU could play a crucial role in
determining the customers’ intention to use such technology. This thought has been
empirically supported by different MB studies (Akturan and Tezcan, 2012; Gu et al.,
2009; Hanafizadeh et al., 2014; Luarn and Lin, 2005; Püschel et al., 2010). In keeping with
the argument of Davis et al. (1989), individuals could also be involved in the cognitive
trade-off process between the efforts required to successfully apply the technology in
front of the benefits and advantages attained by using such technology. Therefore,
Davis et al. (1989) articulated that PEOU could contribute to the BI directly or indirectly
by facilitating the impact of PU on BI. Such a causal relationship between PEOU
and PU has been largely proven by many researchers (e.g. Gu et al., 2009; Luarn and
Lin, 2005) who examined the customer adoption of MB. Thus, this study assumes the
following hypotheses:
H3. PEOU will positively influence Jordanian customers’ intention to adopt MB.
H4. PEOU will positively influence PU of using MB.
3.3 PR
According to Pavlou (2001, p. 109), PR is conceptualised as “the consumer’s subjective
expectation of suffering a loss in pursuit of a desired outcome”. In fact, customers could
experience different kinds of risk such as performance, social, financial, psychological,
and physical which makes the impacting the role of PR on BI more complicated
(Featherman and Pavlou, 2003). Moreover, customers are more apprehensive for the
aspects pertaining to disconnection problems and their probability; this is coupled with
their concerns associated with third parties, electronic piracy, and cybercrimes which, in
turn, lets customers be more hesitant in accepting online banking channels (Poon, 2008).
Indeed, there are several reasons for supporting and including PR among the
conceptual model proposed. First, outcomes of using electronic banking channels have
been extensively characterised with a high degree of uncertainty, intangibility,
heterogeneity, and vagueness (Curran and Meuter, 2005; Eriksson et al., 2008; Gan et al.,
2006; Jaruwachirathanakul and Fink, 2005; Kolodinsky et al., 2004). Therefore, using
such a channel to attain the financial transactions could comprise of further financial,
performance, and privacy risks (Martins et al., 2014). Second, an increase in the rate of
Downloaded by New York University At 01:34 05 February 2016 (PT)
electronic financial crimes in Jordan over the last decade in addition to the media focus on
such crimes represents another reason for the increased concerns among the Jordanian
banking customers to use the electronic banking channels (Addustour, 2012; Al Sukkar
and Hasan, 2005; BBC Arabic, 2009; SarayNews, 2011). Therefore, it was clearly noticed
that PR could be one of the most important aspects that might play a vital role in forming
the Jordanian customers’ intention to use the SST banking channels.
A closer look at the relevant studies leads the authors to observe that PR is one
of the most important obstacles hindering the customers’ willingness to adopt MB
(i.e. Akturan and Tezcan, 2012; Cruz et al., 2010; Hanafizadeh et al., 2014; Lee et al.,
2007). Furthermore, in their meta analyses of 25 articles in the area of online channels,
Taylor and Strutton (2010) asserted a negative impact of PR on the BI with an average
weight effect of ?0.46. Accordingly, this study assumes the following hypothesis:
H5. PR will negatively influence the Jordanian customers’ intention to adopt MB.
3.4 SE
SE could be identified as perception and confidence of individuals in their ability to
manage and conduct a set of particular actions needed to achieve specified kinds of
performances (Bandura, 1986). Within the context of the IS, Compeau and Higgins
(1995) stressed the important role of SE in contributing to both individuals’ willingness
to adopt new technology and their perception towards the expectation of outcomes by
using such systems. These expectations were categorised by Compeau and Higgins
(1995) into two subgroups: performance expectation, which is related to behavioural
outcomes of job performance; and personal outcome expectations such as individual
esteem and sense of accomplishment. Accordingly, it could be argued that banking
customers, who enjoy an adequate level of SE, are more likely to perceive using MB to
be useful in their life as well as the simple technology to be used (Püschel et al., 2010;
Wang et al., 2003). This is especially important when considering the nature of MB as
one of the most recent and novel kind of self-service banking technologies requiring the
customer to conduct financial transactions by himself and away from any support of
banking staff (Püschel et al., 2010; Zhou et al., 2010). However, Venkatesh et al. (2003)
indicated that the conceptual and operational dimensions of SE are different from effort
expectancy (PEOU). Venkatesh et al. (2003) also added that the impact of SE on the BI is
restricted by the mediating effect of PEOU. Therefore, the current study proposes SE
as an indirect predictor of BI by being a mediating role of both PU and PEOU.
Such a role of SE in predicting both PU and PEOU has been commonly approved by
different researchers over the relevant area of interest. For instance, in their study to
examine the adoption of internet banking, Wang et al. (2003) empirically verified a strong
association between SE with both PEOU and PU. Similarly, Zhao et al. (2008) noticed a
significant effect for the SE on PEOU. Cheng et al. (2008) also documented that SE had a
significant influence on effort expectancy and performance expectancy. Furthermore,
PEOU which was related to using internet banking was noticed by Al-Somali et al. (2009)
to be influenced by SE. A study examining the adoption of MB by Luarn and Lin (2005)
empirically validated a strong relationship between SE and PEOU as well.
Further customers’ perception and confidence in their ability to use MB could help
them to mitigate their anxious risks perceived in using such a system (Bandura, 1986;
Luo et al., 2010; Walker and Johnson, 2006; Zhou, 2012). This though has also been
empirically supported by Walker and Johnson (2006) who noticed that observed PR was
negatively correlated with customer beliefs of capacity. Indeed, Walker and Johnson
(2006) argued that customers are more likely to see that using self-service technologies
adoption of
MB in Jordan
Downloaded by New York University At 01:34 05 February 2016 (PT)
will be less risky if they have a positive perception and confidence in their ability to use
such a system. Kim et al. (2005) also supported that PR related to using e-commerce
applications is significantly predicted by the role of SE.
Thus, this study proposes the subsequent hypotheses:
H6. SE will positively influence PU of using MB.
H7. SE will positively influence PEOU related to using MB.
H8. SE will negatively influence PR in using MB.
4. Research methodology
A self-administered questionnaire was employed to collect data using a convenience
sample of 500 Jordanian banking customers from two main cities: Amman and
Al-Balqa’. In total, 28 scale items were adapted from prior IS literature (Compeau
and Higgins, 1995; Davis et al., 1989; Featherman and Pavlou, 2003) to measure the
main underlying constructs encompassed in the conceptual model (see Table AI).
The main constructs of the TAM (PU and PEOU) were measured by items adapted
from Davis et al. (1989) while items for PR were drawn from Featherman and Pavlou
(2003). Featherman and Pavlou’s (2003) scale covered the main dimensions of PR
(i.e. performance risk, financial risk, privacy risk, and social risk). Five items of SE were
selected from Compeau and Higgins’ (1995) scale. All of these items were measured
using the seven-point Likert scale ranging from “strongly disagree (1)” to “strongly
agree (7)” (Dwivedi et al., 2006). As seen in Table AI, five of the most common banking
transactions were also selected to measure the adoption behaviour of MB. The
seven-point time scale was adopted to measure the adoption behaviour towards these
services with anchors including: never, once a year, several times a year, once a month,
several times a month, several times a week, several times a day (Venkatesh et al., 2012).
The questionnaire also included six closed-ended questions to represent the respondents’
demographic characteristics.
As Arabic is the native language of the respondents being targeted in the current
study (Jordanian banking customers), the questionnaire, therefore, was converted into the
Arabic language using the back translation method (Brislin, 1976). This was followed by
conducting a pilot study to assure the adequate level of reliability in the measurements
used as well as to avoid any confusions or contradictions prior to conducting the main
survey. Therefore, thirty questionnaires were allocated to a sample of Jordanian banking
customers who were asked to fill the given questionnaire and provide any comments
about it. Indeed, only twenty-two questionnaires were returned. Noticeably, the vast
majority of those respondents indicated that the language used was clear and filling the
questionnaire did not consume much time. Moreover, an inspection of Cronbach’s ?
values was undertaken to see the scale items being able to have an acceptable level
of reliability (Bhattacherjee, 2012; Sekaran, 2000). Table I presents all the values of
Cronbach’s coefficient ?; they were found to be as low as (0.75) and as high as (0.92).
Therefore, all were able to be above 0.70 as suggested by Nunnally (1978).
5. Results
5.1 Respondents’ profile and characteristics
Out of 500 questionnaires distributed, 343 (68.8 per cent) completed questionnaires were
returned and considered valid for further statistical analyses. The main descriptive
statistics suggested that the vast majority of respondents were male (65.6 per cent) with
Downloaded by New York University At 01:34 05 February 2016 (PT)
their age ranging mainly from 25 to 40 (66 per cent) and most of them have a Bachelor’s
degree or above (80 per cent). About 91 per cent of the respondents also have had more
than three years’ experience with the computer and internet.
5.2 Outliers
An examination of the values of Mahalanobis-D squared distance (D2) provided in the
AMOS output file indicated that there are 17 outlier cases with a p-value less that
cut-off point o0.001 (see Table II). Even though removing these outlier cases could
enhance the multivariate analysis, the results generalisability could be negatively
reflected by doing this (Hair et al., 2006; Tabachnick and Fidell, 2007). Moreover, due to
the large sample size for the three data sets, a small number of outliers would not be
problematic (Kline, 1998; Tabachnick and Fidell, 2007). Accordingly, the decision was
taken to retain these outliers for all three data sets.
5.3 Normality
A skewness-kurtosis approach was adopted to test univariate normality for each variable
(Hair et al., 2010; Kline, 2005). Using AMOS21, the statistical values of skewness and
kurtosis were tested and found that they were within their respective levels. As reported


I'm Natalie

You need a custom essay? I have some suggestions for you...

Check it out