Ijbm-01-2020-0039 proof 1575
Download 334,47 Kb. Pdf ko'rish
|
Mindfulness, Mobile paymen
|
H1c
H1a H1b H2a H2b H3a H3b H4a H4b H5 NS EG LC CN
dimensions: Technological Novelty Seeking (NS), Engagement with the Technology (EG), Awareness of Local Contexts (LC), and Cognizance of Alternative Technologies (CN) Figure 3. Conceptual model of mobile payment adoption IJBM
38,7 1584
questions to confirm that the respondents actually paid attention to the questionnaire, and the consistency of their responses. The respondent profile is presented in Table 2 . The sample included 794 valid responses (380 from Spain and 414 from the USA). Spain was selected as one of the countries with the highest annual growth rates in mobile payment use (
Morgan, 2019 ), having experienced rates significantly above 100% in recent years ( Statista, 2020 ). The United States was taken as a reference as it is the world ’s leading economy and is the developed country most used as a benchmark in the vast majority of economic and financial analyses. In 2020, Spain ranked 16th internationally in mobile payment transaction volume, reaching US$7,423m, while the US ranked second, reaching US$357,557m ( Statista, 2020 ). Smartphone penetration rates are slightly higher in the US than in Spain, at 79.1% and 74.3% respectively ( Newzoo, 2020 ). However, transmission speeds are somewhat higher in Spain. Cellular download and upload speed (mb/s) in the US are 21.3 and 6.3, respectively, and in Spain, 24.8 and 9.3, respectively ( Opensignal, 2019 ). The highest percentage of users in both countries are in the 25 –34 and 35–44 age segments. However, the penetration rate of the youngest segment is significantly higher in the US than in Spain, with values of 34.6% and Frequency (%) Spain
USA Total
Gender Male
60.3 52.2
56.0 Female
38.9 43.5
41.3 Not indicated 0.8 4.3
2.6 Age
<25 years 30.3
30.9 30.6
25 –39 years 43.4 53.4
48.6 40 –54 years 22.4 8.7
15.2 >54 years 3.2 2.7
2.9 Not indicated 0.8 4.3
2.6 Education level Primary school 1.1
2.2 1.6
High school 27.4
17.6 22.3
Colleges and universities 70.8
75.8 73.4
Not indicated 0.8
4.3 2.6
Operating system of mobile devices Android
80.0 47.6
63.1 IOS (Apple) 19.2 47.8
34.1 Other
0 0.2
0.1 Not indicated 0.8 4.3
2.6 Frequency of mobile payment use Occasionally 59.2
47.8 53.3
Frequently 26.6
33.3 30.1
Very frequently 14.2
14.5 14.4
How many times have you used mobile payment? 1 –10 times 53.2 53.1
53.1 >10 times 46.8 46.9
46.9 For how long have you used mobile payment? 0 –6 months 49.7 55.2
52.6 >6 months 50.3 44.8
47.4 Table 2.
Respondent profile Mindfulness in mobile payments
adoption 1585
27.2% in the US, compared to 27.9% and 29.2% in Spain ( Statista, 2020 ). The highest percentages of users are in the lower income segments in both countries, with low, medium and high-income users representing 40.8%, 35.7% and 23.5% in the US, and 40.5%, 31.4% and 28.1 in Spain ( Statista, 2020 ). 3.2 Reliability, dimensionality and validity of the scales First, we verified the initial reliability of all the scales. They were all above the recommended 0.7 for Cronbach ’s alpha ( Nurosis, 1993 ) and 0.3 for the item-total correlation ( Oliver, 1980 ). Second, to confirm the dimensional structure of the scales, a series of confirmatory factor analyses (CFA) were carried out ( Hair et al., 1999 ). Statistical software EQS version 6.4 was used. We used the robust maximum likelihood method, as it operates well in samples that do not unequivocally overcome the multivariate normality test. Third, to purify the model ’s construct items, those that failed to meet any of the three criteria below ( J €oreskog and S€orbom, 1993 ) were eliminated. Weak convergence variables ( Steenkamp and Van Trijp, 1991 ), indicators that did not show significant factorial regression coefficients (t-student > 2.58, p 5 0.01), were eliminated. Strong convergence variables ( Steenkamp and Van Trijp, 1991 ), which included all indicators whose standardized coefficients were less than 0.5, were eliminated ( Hildebrant, 1987 ). Specifically, those indicators whose R 2 is less than 0.5 were excluded to ensure that for each indicator error variance did not exceed 50%. Through this process, five items were deleted (NS2, CN1, CN2, PEO2). The purified measurement items of the convergent validity of the measurement items are presented in Table 4
. Finally, the CFA showed good fit (see Table 3 ). Following verification that the scale and the measurement items had good fit, the validity of the quality of the constructs was tested by examining their convergent and discriminant validity. Convergent validity was verified as the factor load of each indicator was higher than 0.5 at a significance level of 0.01. Similarly, average variance extracted was analyzed using this criterion; measurements of convergent validity should contain less than 50% of the error variance, so the AVE should be greater than 0.5. As seen in Table 4
, all of the variables passed the convergent validity test. Second, with regard to discriminant validity, the square root of the AVE statistic was compared with the inter-construct correlations (values outside the diagonal). In this case, to ensure discriminant validity, the diagonal values must be higher (see Table 5 ). 3.3 Hypothesis testing results The results showed that, for the hypotheses that reflect the direct relationship between mindfulness and other variables, the direct effect was significant on perceived usefulness [ β 5 0.889; p < 0.05], perceived ease of use [β 5 0.705; p < 0.05], and attitude [β 5 0.588; p < 0.01]; therefore, hypotheses H1a
, H1b
and H1c
are supported. Hypotheses 2a and 2b
γ 5 0.601; p < 0.05] and the user’s intention to use mobile payment [ γ 5 0.812; p < 0.05], which were shown to be significant. Recommended value CFA Absolute fit adjustment RMSEA RMSEA < 0.08 0.047 90% Confidence Interval of RMSEA (0.043; 0.050) Incremental fit Adjustment NFI NFI > 0.9 0.941 NNFI
NNFI > 0.9 0.954
CFI Near of 1 0.962 IFI
Near of 1 0.962
Table 3. Confirmatory factor analysis IJBM
38,7 1586
Variable Item
t-student St. coeff R 2
FCC Mindfulness Technological novelty seeking 0.69
0.82 NS1
30.58 0.845
0.715 NS3
26.61 0.816
0.666 Engagement with the technology 0.60 0.82
EG1 24.73
0.783 0.613
EG2 26.38
0.807 0.651
EG3 23.62
0.730 0.533
Awareness of local contexts 0.62
0.83 LC1
25.10 0.762
0.580 LC2
28.13 0.819
0.670 LC3
23.39 0.775
0.601 Cognizance of alternative technologies 0.73 0.84
CN3 24.35
0.842 0.710
CN4 25.29
0.868 0.753
Perceived ease of use 0.73
0.89 PEO1
25.20 0.818
0.669 PEO3
28.78 0.895
0.802 PEO4
27.33 0.856
0.716 Perceived usefulness 0.72 0.91
PUS1 28.51
0.815 0.664
PUS2 32.27
0.857 0.735
PUS3 36.16
0.891 0.793
PUS4 29.25
0.827 0.684
Intention to use 0.74
0.89 IU1
26.76 0.838
0.702 IU2
35.09 0.844
0.713 IU3
36.42 0.895
0.800 Attitude
0.77 0.93
ATT1 33.13
0.906 0.821
ATT2 28.13
0.847 0.717
ATT3 28.24
0.893 0.798
ATT4 26.75
0.855 0.731
Subjective norms 0.82
0.93 SN1
36.03 0.893
0.797 SN2
39.38 0.924
0.854 SN3
37.55 0.902
0.813 Construct F1 F2
F4 F5 F6 F7 F8 F9 F1: Technological novelty seeking 0.83
F2: Engagement with the technology 0.647 0.77
F3: Awareness of local contexts 0.585
0.664 0.79
F4: Cognizance of alternative technologies 0.473 0.556
0.618 0.85
F5: Perceived ease of use 0.472
0.267 0.301
0.267 0.85
F6: Perceived usefulness 0.505
0.416 0.506
0.329 0.636
0.85 F7: Intention to use 0.578 0.420
0.450 0.288
0.658 0.636
0.86 F8: Attitude 0.532 0.416
0.443 0.337
0.695 0.752
0.705 0.88
F9: Subjective norms 0.400
0.362 0.367
0.213 0.277
0.596 0.614
0.540 0.90
Note(s): Italic values represents the square root of the AVE statistic Table 4.
Quality criterion and factor loadings Table 5. Analysis of discriminant validity Mindfulness in mobile payments
adoption 1587
Perceived ease of use had a significant and positive effect on perceived usefulness [ γ 5 0.712; p < 0.05] and attitude [ γ 5 1.007; p < 0.05], validating hypotheses 3a and
3b . H4a and H4b
were also supported, as subjective norms proved to have a significant effect on attitude [ β 5 0.864; p < 0.05] and intention to use [ β 5 0.685; p < 0.05]. Attitude was shown to directly and positively influence intention to use [ γ 5 0.872; p < 0.05], supporting hypothesis H5 . Therefore, all the hypotheses were well supported. Finally, the model showed good fit (RMSEA
5 0.043; NFI 5 0.943; NNFI 5 0.960; IFI 5 0.965; CFI 5 0.965) (see Figure 4
). 3.4 Multisample analysis In order to assess the generalization capacity of the results a multisample analysis was performed. This type of analysis allows us to test for possible differences between two or more groups. In our case, the sample was divided into 2 groups based on the respondent ’s nationality (Spanish or USA). First, the multisample analysis offered a structural solution for each group (see Table 6
). The results did not show any differences regarding the test of the Mindfulness Perceived usefulness Perceived ease of use Subjective norms
Attitude Intention to use 0,588*
0.705* 0.889*
0.601* 0.812*
0.712* 1.007*
0.864* 0.685*
0.872* NS EG LC CN 0.991* 1.057* 0.952*
0.821* Download 334,47 Kb. Do'stlaringiz bilan baham: 
|