CHAPTER 5 DATA ANALYSIS AND RESULTS
5.1. Validation of measures: reliability and validity
5.1.1. First-order and reflective – reflective second-order constructs
In this study, the reliability of the reflective measurements were assessed by Cronbach’s Alpha, CR (composite reliability) and AVE (average variance extracted).
The analysis results showed that except CFCF4 and CFCI4 were rejected because of low loadings, the loadings of most of the indicators of the constructs were above the 0.70. Also, Cronbach’s Alpha and CR were greater than 0.70.
Table 5-11: Cronbach's Alpha and composite reliability Constructs and indicators Mean Cronbach’s
Alpha
Composite reliability
CFC-Immediate (CFCI) 4.70 0.89 0.92
CFC-Future (CFCF) 5.10 0.90 0.92
Financial Risk (FR) 5.57 0.87 0.91
Performance Risk (PER) 5.38 0.85 0.90
Privacy Risk (PrR) 5.34 0.79 0.88
Psychological Risk (PSR) 5.39 0.80 0.91
Social Risk (SR) 5.72 0.82 0.92
Time Risk (TR) 5.71 0.86 0.90
Perceived Confidentiality
(PC) 4.72 0.90 0.92
Perceived Availability (PA) 4.29 0.77 0.87
Perceived Non-Repudiation
(PNR) 5.29 0.89 0.93
Continuance intention to use
mobile commerce (UMC) 3.79 0.88 0.92
Perceived Risk (PR) - 0.78 0.85
(Source: author’s calculation) Finally, all of AVEs were close or greater than 0.5. Thus, the convergent validity of all constructs was acceptable. Please also see Table 5-2.
Table 5-12: Factor loadings and average variance extracted
Constructs and indicators Factor
loadings
Average variance extracted CFC-Immediate (CFCI)
(Joireman et al., 2012, Strathman et al., 1994)
0.65 I only use mobile commerce to satisfy immediate
concerns, figuring the future will take care of itself (CFCI1)
0.79
My mobile commerce activities are only influenced 0.84
Constructs and indicators Factor loadings
Average variance extracted by the immediate (i.e., a matter of days or weeks)
outcomes of my actions (CFCI2)
My convenience is a big factor in my mobile commerce activities (CFCI3)
0.77 I think that sacrificing mobile commerce activities
now is usually unnecessary since future outcomes can be dealt with at a later time (CFCI5)
0.83
I only use mobile commerce to satisfy immediate concerns, figuring that I will take care of future problems that may occur at a later date (CFCI6)
0.85
Since my day-to-day mobile commerce activities have specific outcomes, it is more important to me than mobile commerce activities that have distant outcomes (CFCI7)
0.75
I generally ignore warnings about possible future problems of mobile commerce activities because I think the problems will be resolved before they reach crisis level (CFCI4) (RI)
-
CFC-Future (CFCF)
(Joireman et al., 2012, Strathman et al., 1994)
0.66 I consider how mobile commerce’s benefits might
be in the future, and try to archive those benefits with my day-to-day of using mobile commerce (CFCF1)
0.84
Often, I engage in a mobile commerce activity in order to achieve outcomes that may not result for many years (CFCF2)
0.86
I am willing to sacrifice my immediate happiness or well-being of using mobile commerce activities in order to achieve future outcomes (CFCF3)
0.80
Constructs and indicators Factor loadings
Average variance extracted I think it is more important to make a mobile
commerce decision with important distant consequences than a mobile commerce decision with less important immediate consequences (CFCF5)
0.81
When I make a mobile commerce decision, I think about how it might affect me in the future (CFCF6)
0.81 My mobile commerce activities are generally
influenced by future consequences (CFCF7)
0.75 I think it is important to take warnings about
negative outcomes of mobile commerce activities seriously even if the negative outcome will not occur for many years (CFCF4) (RI)
-
Financial Risk (FR) (Featherman and Pavlou, 2003, Kim et al., 2005)
0.71 Mobile commerce would be an inappropriate way
to spend my money (FR1)
0.85 The money I would make on mobile commerce
would not be wise (FR2)
0.83 I will not get my money’s worth from mobile
commerce (FR3)
0.82 Mobile commerce would not provide value for the
money I spent (FR4)
0.88 Performance Risk (PER) (Featherman and Pavlou,
2003, Kim et al., 2005)
0.69 I worry mobile commerce will not perform as they
are supposed to (PER1)
0.85 I worry mobile commerce will not provide the level
of benefits as I expect (PER2)
0.83 A lot of risks would be involved with purchasing 0.86
Constructs and indicators Factor loadings
Average variance extracted items on mobile commerce (PER3)
I am not confident about mobile commerce vendors to perform as expected (PER4)
0.78 Privacy Risk (PrR) (Featherman and Pavlou,
2003, Kim et al., 2005)
0.70 Using mobile commerce, I will lose control over
my payment information (PR1)
0.82 Using mobile commerce, my personal information
would be used without my knowledge (PR2)
0.86 Internet criminals might take control of my account
if I use mobile commerce (PR3)
0.84 Psychological Risk (PSR) (Featherman and
Pavlou, 2003, Kim et al., 2005)
0.83 Mobile commerce will not fit in well with my self-
image or self-concept (PSR1)
0.92 The usage of mobile commerce would lead to a
psychological loss for me (PSR2)
0.90 Social Risk (SR) (Featherman and Pavlou, 2003,
Kim et al., 2005)
0.85 Mobile commerce will negatively affect the way
others think of you (SR1)
0.90 Using mobile commerce would lead to a social loss
for me (SR2)
0.94 Time Risk (TR) (Featherman and Pavlou, 2003,
Kim et al., 2005)
0.70 Using mobile commerce, I will lose time switching
to a different payment method (TR1)
0.83 Using mobile commerce, I will waste a lot of time
fixing payments errors (TR2)
0.87 Mobile commerce could lead to inefficient use of
my time (TR3)
0.87
Constructs and indicators Factor loadings
Average variance extracted Mobile commerce will take too much time or be a
waste of time (TR4)
0.78 Perceived Confidentiality (PC) (Hartono et al.,
2014) 0.66
Someone uses my mobile commerce ID to read my transactional informationR (PC1)
0.78 Someone uses my mobile commerce ID to make
order R (PC1)
0.84 Someone steals my mobile commerce ID R (PC1) 0.89 The site transmits my transactional information
accurately(PI1)
0.81 My transactional information is alteredR (PI2) 0.85 The site records my transactional information
incorrectlyR (PI3)
0.69
Perceived Availability (PA) (Hartono et al., 2014) 0.69 I cannot order due to system failureR (PA1) 0.82
I cannot order due to database failureR (PA2) 0.82 I cannot order due to network failureR (PA3) 0.84 Perceived Non-Repudiation (PNR) (Hartono et al.,
2014)
0.83 The site uses a digital signature (PNR1) 0.90
The legislation backs up the digital signature (PNR2)
0.93 The identity of this site is trustworthy (PNR3) 0.89 Continuance intention to use mobile commerce
(UMC) (Chong, 2015)
0.80 I intend to increase my use of mobile commerce in
the future (UMC1)
0.87 I intend to continue my use of mobile commerce in
the future (UMC2)
0.91 I will strongly recommend others to use mobile
commerce (UMC3)
0.91
Constructs and indicators Factor loadings
Average variance extracted
Perceived Risk (PR) 0.48
FR 0.68
PER 0.76
PrR 0.73
PSR 0.67
SR 0.59
TR 0.71
Note: (RI) Rejected items in the accuracy test
(Source: author’s calculation) Fornell and Larcker (1981) criterion and the Heterotrait-Monotrait (HTMT) correlations matrix were used to assess the discriminant validity of reflective constructs. Following Fornell and Larcker (1981) criterion, we compared the square root of the AVE value of each construct with the highest bivariate correlations with other constructs. The results indicated that the square root of each AVE was greater than its highest bivariate correlations, and thus demonstrated the discriminant validity was acceptable. Furthermore, the results indicated that all HTMT values were below 0.85 (Hair et al., 2017), thus, consolidating the discriminant validity of studied constructs. Please also see Table 5-3.
Table 5-13: Fornell-Larcker and Heterotrait-Monotrait criterion
1 2 3 4 5 6 7 8 9 10 11 12 13
1.CFC-I 0.81 0.36 0.39 0.45 0.41 0.28 0.30 0.28 0.25 0.31 0.31 0.18 0.51 2.CFC-F 0.33 0.81 0.13 0.19 0.10 0.21 0.18 0.27 0.46 0.35 0.42 0.24 0.25 3.FR 0.34 0.11 0.85 0.51 0.34 0.43 0.29 0.46 0.07 0.16 0.13 0.10 0.81 4.PER 0.40 0.17 0.45 0.83 0.44 0.54 0.43 0.44 0.10 0.23 0.15 0.08 0.86 5.PrR 0.34 0.08 0.29 0.36 0.84 0.40 0.30 0.58 0.10 0.20 0.13 0.07 0.78 6.PSR 0.24 0.18 0.36 0.45 0.32 0.91 0.43 0.43 0.09 0.23 0.12 0.09 0.76 7.SR 0.26 0.16 0.25 0.36 0.25 0.35 0.92 0.34 0.05 0.14 0.10 0.04 0.63 8.TR 0.25 0.24 0.40 0.38 0.48 0.36 0.29 0.84 0.16 0.39 0.25 0.05 0.85 9.PC 0.23 0.42 0.04 0.08 0.08 0.07 0.04 0.13 0.81 0.67 0.70 0.29 0.13 10.PA 0.26 0.30 0.13 0.18 0.16 0.19 0.12 0.32 0.56 0.83 0.65 0.27 0.33 11.PNR 0.29 0.38 0.12 0.13 0.11 0.10 0.09 0.22 0.63 0.55 0.91 0.17 0.22 12.UMC 0.16 0.23 -
0.08 - 0.05
- 0.03
-0.07 0.00 -0.02 0.27 0.23 0.16 0.90 0.10
13.PR 0.45 0.23 0.71 0.77 0.64 0.65 0.53 0.74 0.11 0.28 0.20 -0.07 0.59 Notes. The square roots of AVE, Pearson correlations and HTMT ratios are on, at the lower, at the upper the diagonal, respectively. CFCF: CFC-Future; CFCI: CFC- Immediate; FR: Financial risk; PER: Performance risk; PrR: Privacy risk; PSR:
Psychological risk; SR: Social risk; TR: Time risk; PC: Perceived confidentiality; PA:
Perceived availability; PNR: Perceived non-repudiation; UMC: Continuance intention to use mobile commerce; PR: Perceived risk. Perceived security was not included in this test due to the reflective – formative nature of this construct (D'Arcy et al., 2009)
(Source: author’s calculation) It is worthy to note that we cannot establish discriminant validity between overall perceived risk and particular perceived risk components as the measurement model of the overall perceived risk repeats the indicators of its particular perceived risk components.