6.1 Demographic Characteristics
Male and female youth from the Sultanate of Oman made up the sample for the poll, with female replies accounting for 65% of the total. The respondents' ages ranged from 19 to 40, with the age group of 20 to 25 accounting for the largest share. Additionally, the bulk of survey participants had bachelor's degrees. The majority of replies are from A' Sharqiyah University,
Table 1. Demographic Characteristics
Demographics
|
No.
|
%
|
Gender
|
Male
|
131
|
35.0
|
|
Female
|
243
|
65.0
|
Age
|
19 years and below
|
10
|
2.7
|
|
20-25
|
219
|
58.6
|
|
26-29
|
42
|
11.2
|
|
30-34
|
50
|
13.4
|
|
35-39
|
26
|
7.0
|
|
40 and above
|
27
|
7.2
|
Marital status
|
Single
|
241
|
64.4
|
|
Married
|
133
|
35.6
|
Academic qualifications
|
High School
|
21
|
5.6
|
|
Diploma
|
106
|
28.3
|
|
Bachelor
|
224
|
59.9
|
|
Postgraduates
|
23
|
6.1
|
University / College
|
A'Sharqiyah University
|
249
|
66.6
|
|
Sultan Qaboos University
|
6
|
1.6
|
|
Sur University College
|
11
|
2.9
|
|
The University of Technology and Applied Sciences
|
26
|
7
|
|
Other
|
82
|
21.5
|
Future targeting Job
|
Government employee
|
207
|
55.3
|
|
Private sector employee
|
87
|
23.3
|
|
Free Business
|
59
|
15.8
|
|
Other
|
21
|
5.6
|
6.2 Reliability and Validity of measurement model
Table (2) shows the assessment of the validity and reliability of the proposed conceptual model. The reliability and validity test results indicate that the factor loading of all items shows a satisfactory level of internal reliability, ranging between 0.603 to 0.765. However, the metrics of construct reliability and validity show reasonable composite reliability, satisfactory internal consistency, and a high level of convergent validity based on the recommended standards (α > 0.7, rho_A > 0.7, CR > 0.7, AVE > 0.5) suggested by (Hair et al., 2014; Henseler et al., 2014).
Table 2. Reliability and Validity
Variable
|
Standardised loadings
|
α
|
rho_A
|
CR
|
AVE
|
Social Perception
|
0.636
|
0.641
|
0.637
|
0.77
|
0.401
|
|
0.637
|
|
|
|
|
|
0.661
|
|
|
|
|
|
0.604
|
|
|
|
|
|
0.626
|
|
|
|
|
Cost of Migrants
|
0.700
|
0.647
|
0.653
|
0.791
|
0.488
|
|
0.765
|
|
|
|
|
|
0.711
|
|
|
|
|
|
0.610
|
|
|
|
|
Adoption of digital economy
|
0.660
|
0.703
|
0.702
|
0.808
|
0.457
|
|
0.708
|
|
|
|
|
|
0.643
|
|
|
|
|
|
0.698
|
|
|
|
|
|
0.666
|
|
|
|
|
Unemployment of Youths
|
0.697
|
0.771
|
0.774
|
0.839
|
0.466
|
|
0.736
|
|
|
|
|
|
0.634
|
|
|
|
|
|
0.711
|
|
|
|
|
|
0.653
|
|
|
|
|
|
0.661
|
|
|
|
|
6.3 Structural Equation Model Assessment
The model's discriminant validity was evaluated using the Fornell-Larcker criteria and the Heterotrait-Monotrait ratio test with the PLS Algorithm. Hence, according to the results of the Fornell-Larcker test, the correlation between the square root of the average variance derived from the construct and itself within the same column is more significant than the correlations between the construct and other variables, with these values demonstrating satisfactory discriminant validity, as shown in Table (3). This demonstrates a relevant difference in the measures that were adopted. Additionally, the results obtained, as shown in Table (4), reveal acceptable discriminant validity, as evidenced by the correlation below the 0.85 level advised by Henseler et al. (2014). As a result, the research analysis developed significant and trustworthy relationships that opened the door for more research. Figure (2) represents the structural model analysis, which highlights the significant explanatory power and reliability of the model construction as evidenced by the R square findings for unemployment of youth (R =519).
Table 3. Discriminant Validity: Fornell-Larcker Criterion
Note: CM: Cost of Migrant, ADE: Adoption of digital economey, SP: Social perception, UOY: Unemployment of Youths
|
CM
|
MQR
|
SP
|
UOY
|
CM
|
0.699
|
|
|
|
ADE
|
0.380
|
0.676
|
|
|
SP
|
0.528
|
0.426
|
0.633
|
|
UOY
|
0.475
|
0.651
|
0.471
|
0.683
|
Table 4. Discriminant Validity: Heterotrait-Monotrait Ratio (HTMT)
Note: CM: Cost of Migrant, ADE: Adoption of digital economy, SP: Social perception, UOY: Unemployment of Youths
|
CM
|
MQR
|
SP
|
UOY
|
CM
|
|
|
|
|
ADE
|
0.560
|
|
|
|
SP
|
0.789
|
0.612
|
|
|
UOY
|
0.669
|
0.872
|
0.624
|
|
6.4 Measurement Correlation
VIF Test
The VIF test was performed to measure the extent to which variance is inflated. Table (6) displays the findings of the variance inflation factor VIF. The results show a moderate correlation between the independent variables, ranging from 1.117 to 1.644.
Table (5). VIF
|
VIF
|
VIF
|
SP01
|
1.349
|
2.349
|
SP02
|
1.397
|
2.397
|
SP03
|
1.117
|
2.117
|
SP04
|
1.252
|
2.252
|
SP05
|
1.316
|
2.316
|
CM01
|
1.293
|
2.293
|
CM02
|
1.366
|
2.366
|
CM03
|
1.241
|
2.241
|
CM04
|
1.144
|
2.144
|
ADE01
|
1.299
|
2.299
|
ADE02
|
1.351
|
2.351
|
ADE03
|
1.235
|
2.235
|
ADE04
|
1.396
|
2.396
|
ADE05
|
1.241
|
2.241
|
UOY01
|
1.51
|
2.51
|
UOY02
|
1.644
|
2.644
|
UOY03
|
1.275
|
2.275
|
UOY04
|
1.494
|
2.494
|
UOY05
|
1.361
|
2.361
|
UOY06
|
1.419
|
2.419
|
Hypothesis Testing
Direct Effect
Bootstrapping of 5000 sub-samples was performed to investigate the suggested interactions between the constructs. The analysis reveals an obvious and significant link between social perception and unemployment of youth as well as a positive relationship between the cost of migrants and unemployment of youth with (t-value = 2.986, p= 0.003) (t-value = 0.047, p= 0.000), respectively. Thereby H1 and H2 are significantly supported. Furthermore, the analysis reveals a very favourable direct relationship between the adoption of digital economy and youth unemployment where (t-value =11.599, p= 0.000); hence H3 is highly supported. However, considering gender as a moderator, significant correlations between social perception and unemployment of youth when gender have moderate effect with (t-value= 2.212, p= 0.027), providing strong support of H4. While insignificant moderate effect of gender between the relationship of cost of migrant, adoption of digital economy with unemployment of Omani Youth’s (t-value= 1.156, p= 0.248) (t-value= 0.938, p= 0.348) respectively. Therefore, H5 and H6 are not supported. Figure 3 depicts the revealed results of bootstrapping test.
Table (6). Direct Effect Testing
Hypotheses
|
Path
|
Original Sample (O)
|
Original Sample (O)
|
Standard Deviation
|
T Values
|
P Values
|
Decision
|
H1
|
SP -> UOY
|
0.142
|
1.142
|
0.048
|
2.986
|
0.003
|
Supported
|
H2
|
CM -> UOY
|
0.201
|
1.201
|
0.047
|
4.313
|
0.000
|
Supported
|
H3
|
ADE -> UOY
|
0.513
|
1.513
|
0.044
|
11.599
|
0.000
|
Supported
|
H4
|
SP->G-> UOY
|
-0.104
|
0.896
|
0.047
|
2.212
|
0.027
|
Supported
|
H5
|
CM->G-> UOY
|
-0.051
|
0.949
|
0.044
|
1.156
|
0.248
|
Not Supported
|
H6
|
ADE->G-> UOY
|
0.060
|
1.060
|
0.064
|
0.938
|
0.348
|
Not Supported
|