variances. It shows how much on an average is explained
by items of other factors.
• Maximum Shared Squared Variance (MSV):
The MSV in the model indicates how well is the factor
explained by items outside the factor (i.e. items of other
constructs).
It was noted that the CR for the habit construct (H) is
<0.70; the AVE for H is also <0.50. Hence, there were
validity concerns with the construct ‘habit’ (H). To achieve a
better model fit, the construct habit (H) was dropped from
the model and the test was repeated. After dropping H, the
revised model produced a good model fit with (v
2 /df) = 1.29
and the RMSEA = 0.031.
7.5.2 Structural Path Model
The structural path model in the next stage provides the causal
relations between independent variables and the dependent
variables. Structural equation modeling has been extensively
used in various fields, such as consumer behavior and marketing research because it offers many advantages over other
procedures. SEM can investigate and correct unreliable measures when numerous indicators of each factor are available.
The main benefit of using SEM is that it can examine the
comprehensive theoretical frameworks in which the influences
of factors are established across multiple levels of variables
(Baumgartner & Homburg, 1996). The SEM is a way of a
multi-variate statistical procedure that can assess the underlying
Table 7.5 Overall CFA−summary of path coefficient of questionnaire items
Paths
Regression Weights
S.E.
C.R.
P-Value
Estimate
Stand
Estimate
PE5
<--PE
1.000
.785
PE4
<--PE
1.005
.734
.065
15.453
***
PE3
<--PE
1.247
.829
.078
15.896
***
PE2
<--PE
1.236
.897
.071
17.469
***
PE1
<--PE
1.055
.863
.063
16.705
***
EE5
<--EE
1.000
.828
EE4
<--EE
.998
.832
.057
17.462
***
EE3
<--EE
1.093
.856
.060
18.227
***
EE2
<--EE
1.108
.816
.065
16.962
***
EE1
<--EE
1.135
.856
.062
18.212
***
SI5
<--SI
1.000
.808
SI4
<--SI
.904
.773
.052
17.291
***
SI3
<--SI
.938
.787
.062
15.226
***
SI2
<--SI
1.009
.854
.060
16.934
***
SI1
<--SI
1.061
.865
.062
17.196
***
FC4
<--FC
1.000
.773
FC3
<--FC
.989
.766
.063
15.606
***
FC2
<--FC
1.073
.839
.072
14.966
***
FC1
<--FC
1.102
.857
.072
15.219
***
M3
<--M
1.000
.857
M2
<--M
.925
.891
.049
19.009
***
M1
<--M
.840
.834
.048
17.662
***
BI3
<--BI
1.000
.830
BI2
<--BI
1.101
.867
.063
17.553
***
BI1
<--BI
1.067
.818
.065
16.354
***
UB1
<--- UB
1.000
.841
UB2
<--- UB
1.032
.807
.063
16.399
***
UB3
<--- UB
1.101
.903
.061
18.189
***
H3
<--H
1.000
.558
H2
<--H
.929
.474
.183
5.076
***
H1
<--H
1.160
.615
.216
5.379
***
Source SPSS/AMOS output
66
7 Empirical Evidence of LMS Adoption in the Middle East
by items of other factors.
• Maximum Shared Squared Variance (MSV):
The MSV in the model indicates how well is the factor
explained by items outside the factor (i.e. items of other
constructs).
It was noted that the CR for the habit construct (H) is
<0.70; the AVE for H is also <0.50. Hence, there were
validity concerns with the construct ‘habit’ (H). To achieve a
better model fit, the construct habit (H) was dropped from
the model and the test was repeated. After dropping H, the
revised model produced a good model fit with (v
2 /df) = 1.29
and the RMSEA = 0.031.
7.5.2 Structural Path Model
The structural path model in the next stage provides the causal
relations between independent variables and the dependent
variables. Structural equation modeling has been extensively
used in various fields, such as consumer behavior and marketing research because it offers many advantages over other
procedures. SEM can investigate and correct unreliable measures when numerous indicators of each factor are available.
The main benefit of using SEM is that it can examine the
comprehensive theoretical frameworks in which the influences
of factors are established across multiple levels of variables
(Baumgartner & Homburg, 1996). The SEM is a way of a
multi-variate statistical procedure that can assess the underlying
Table 7.5 Overall CFA−summary of path coefficient of questionnaire items
Paths
Regression Weights
S.E.
C.R.
P-Value
Estimate
Stand
Estimate
PE5
<--PE
1.000
.785
PE4
<--PE
1.005
.734
.065
15.453
***
PE3
<--PE
1.247
.829
.078
15.896
***
PE2
<--PE
1.236
.897
.071
17.469
***
PE1
<--PE
1.055
.863
.063
16.705
***
EE5
<--EE
1.000
.828
EE4
<--EE
.998
.832
.057
17.462
***
EE3
<--EE
1.093
.856
.060
18.227
***
EE2
<--EE
1.108
.816
.065
16.962
***
EE1
<--EE
1.135
.856
.062
18.212
***
SI5
<--SI
1.000
.808
SI4
<--SI
.904
.773
.052
17.291
***
SI3
<--SI
.938
.787
.062
15.226
***
SI2
<--SI
1.009
.854
.060
16.934
***
SI1
<--SI
1.061
.865
.062
17.196
***
FC4
<--FC
1.000
.773
FC3
<--FC
.989
.766
.063
15.606
***
FC2
<--FC
1.073
.839
.072
14.966
***
FC1
<--FC
1.102
.857
.072
15.219
***
M3
<--M
1.000
.857
M2
<--M
.925
.891
.049
19.009
***
M1
<--M
.840
.834
.048
17.662
***
BI3
<--BI
1.000
.830
BI2
<--BI
1.101
.867
.063
17.553
***
BI1
<--BI
1.067
.818
.065
16.354
***
UB1
<--- UB
1.000
.841
UB2
<--- UB
1.032
.807
.063
16.399
***
UB3
<--- UB
1.101
.903
.061
18.189
***
H3
<--H
1.000
.558
H2
<--H
.929
.474
.183
5.076
***
H1
<--H
1.160
.615
.216
5.379
***
Source SPSS/AMOS output
66
7 Empirical Evidence of LMS Adoption in the Middle East