2 THE RELIABILITY TEST: THE CRONBACH’S ALPHA TEST
Tải bản đầy đủ
PERSONAL VALUEATTITUDEENVIRONMENTAL COMMITMENT
24
The results of Cronbach’s alpha test for each construct were summarized in below
table:
Table 4.2:
The results of Cronbach’s alpha
Scale Mean
Item Deleted
if Scale
Corrected
Variance
if ItemTotal
Item Deleted Correlation
CV1
15.100
8.668
0.701
CV2
14.787
8.907
0.713
CV3
14.693
8.979
0.696
CV4
15.067
8.935
0.709
CV5
14.993
9.107
0.665
Collectivistic Values
Cronbach’s alpha
IV1
6.740
4.690
0.664
IV2
6.727
5.274
0.627
IV3
6.573
4.985
0.660
Individualistic Values
Cronbach’s alpha
ICP1
6.473
5.177
0.719
ICP2
6.573
5.092
0.754
ICP3
6.820
5.115
0.724
Individual Consequences of Purchase Cronbach’s alpha
ECP1
9.827
7.433
0.708
ECP3
10.187
6.878
0.750
ECP3
10.327
7.040
0.762
ECP4
10.240
6.573
0.763
Environmental
Consequences
of Cronbach’s alpha
Purchase
EC1
10.973
5.637
0.694
EC2
10.933
5.499
0.774
EC3
10.867
5.781
0.703
EC4
10.867
5.740
0.772
Environmental Commitment
Cronbach’s alpha
Cronbach’s
alpha if Item
Delected
0.843
0.840
0.844
0.841
0.851
0.871
0.717
0.755
0.720
0.803
0.813
0.781
0.809
0.858
0.863
0.847
0.842
0.843
0.882
0.859
0.826
0.854
0.828
0.877
The result performed that five scales had the result of Cronbach’s alpha above
0.6. The highest was 0.882 (environmental consequences of the purchase), and the
lowest was 0.803 (individualistic values). Moreover, the corrected itemtotal
correlation of each item is above 0.4. This indicates that all scales fit the requirement
PERSONAL VALUEATTITUDEENVIRONMENTAL COMMITMENT
of reliability. Therefore, these measures were used in establishing the main survey to
test to the study hypotheses.
4.3 EXPLORATORY FACTOR ANALYSIS (EFA)
After analyzing the Cronbach’s alpha, the author evaluated the measurement
scales by conducting exploratory factor analysis. The purpose of EFA is to define
which set of items go together as a group or are answered similarly by respondents
(Leech et al., 2005). In this study, EFA was run through the Principal Axis Factoring
with Varimax rotation method. There are five factors in the conceptual model:
collectivistic values, individualistic values, individual consequences of purchase,
environmental consequences of purchase, environmental commitment. The author
examined if the items belonging to one concept actually are in the same group.
Based on the test of assumption, the KMO was 0.921 presenting sufficient
items for each factor. The Bartlett test was significant (0.000 less than 5%); this means
that the variable are correlated highly enough to provide a reasonable basic factor
analysis.
Table 4.3:
KMO and Bartlett's Test
KaiserMeyerOlkin Measure of Sampling
Adequacy.
Bartlett's Test of
Approx. ChiSquare
Sphericity
df
Sig.
.921
1717.631
171
.000
By doing EFA (Principal Axis Factoring with Varimax rotation method), the
result showed that five factors were extracted from 19 items measuring: collectivistic
values, individualistic value, individual consequences of purchase, environmental
consequences of purchase, environmental commitment. In addition, the cumulative of
25
PERSONAL VALUEATTITUDEENVIRONMENTAL COMMITMENT
26
the first five factors occupied for 64.166 percent of variance. This indicated that
nearly 64 percent of variance could be explained by five initial items.
Table 4.4:
Total Variance Explained
Factor
Extraction Sums of Squared
Rotation Sums of Squared
Loadings
Loadings
Initial Eigenvalues
% of
Cumulative
% of
Total Variance
Cumulative
Total Variance
Cumulative
Total Variance
%
1
8.677
45.668
45.668 8.324
43.808
43.808 3.034
15.970
15.970
2
2.015
10.604
56.272 1.656
8.714
52.522 2.581
13.583
29.553
3
1.331
7.007
63.280
.995
5.239
57.761 2.551
13.428
42.982
4
1.126
5.927
69.206
.763
4.013
61.774 2.053
10.803
53.785
5
.817
4.301
73.508
.454
2.392
64.166 1.972
10.381
64.166
6
.634
3.335
76.843
7
.588
3.095
79.938
8
.485
2.552
82.490
9
.432
2.275
84.765
10
.405
2.132
86.897
11
.356
1.873
88.769
12
.333
1.755
90.524
13
.310
1.633
92.158
14
.300
1.582
93.739
15
.277
1.456
95.195
16
.258
1.359
96.554
17
.235
1.237
97.791
18
.227
1.197
98.988
19
.192
1.012
100.000
Extraction Method: Principal Axis Factoring.
%
% of
%
PERSONAL VALUEATTITUDEENVIRONMENTAL COMMITMENT
27
The Rotated Factor Matrix showed the items and factor loading for rotated
factors with loading higher than 0.6 are significant as requirement. The items clustered
into five groups that they belong to.
Table 4.5:
Rotated Factor Matrix
Factor
Collectivistic
Values
Individualistic
Values
Individual
consequences of
purchase
Environmental
consequences of
purchase
Environmental
Commitment
CV1
CV2
CV3
CV4
CV5
IV1
IV2
IV3
ICP1
ICP2
ICP3
ECP1
ECP2
ECP3
ECP4
EC1
EC2
EC3
EC4
1
0.684
0.673
0.630
0.672
0.628
2
Component
3
4
5
0.667
0.617
0.701
0.655
0.741
0.643
0.664
0.628
0.675
0.705
0.617
0.757
0.711
0.735
4.4 MULTIPLE REGRESSION ANALYSIS
After testing the Cronbach’s alpha analysis and EFA, the author conducted the
multiple regression analysis in order to define the relationship between five factors
mention above. The purpose of multiple regression analysis is to help the researcher to
predict the level of impact of independent variable on dependent variable (Hair et al.,
2010). Basing on the proposed conceptual model, the researchers ran three times of
multiple regression analysis. The first time was to define the impact of collectivistic
PERSONAL VALUEATTITUDEENVIRONMENTAL COMMITMENT
values and individualistic values on the attitude toward the individual consequences of
the purchase. The second time was to evaluate the impact of collectivistic values and
individualistic values on the attitude toward the environmental consequences of the
purchase. The last time was to justify the influence of the attitude toward the
individual consequences of the purchase and the environmental consequences of the
purchase on environmental commitment.
It is necessary to test these following main assumptions in order to make sure
the multiple regression preforms exactly analysis result:

Assumption 1: There was the linear relationship independent variables and
dependent variable.

Assumption 2: The residual or error was distributed normally.

Assumption 3: No multicollinearity. This assumption is important because
multicollinearity could lead to misleading and/ or inaccurate results.
According to Leech et al., (2005), when there are high intercorrelations
among some set of the predictor variables, multicollinearity occurs.
Multicollinearity could be checked from a correlation matrix and
coefficients results.
4.4.1 Multiple regression analysis: to define the impact of collectivistic values and
individualistic values on the attitude toward the individual consequences of the
purchase.
Results of testing assumptions
Assumption 1:
This assumption could be checked by using Curve Estimation Graphs and
Scatter Graphs to define the relationship between the individual consequences of the
purchase and each predictor. As the plot shown in Appendix H, the relationship
28
PERSONAL VALUEATTITUDEENVIRONMENTAL COMMITMENT
between the individual consequences of the purchase and each predictor was linear
regression. Therefore, this assumption was satisfied.
Assumption 2:
The normal distribution of dependent variable was test by histogram and scatter
plot, which shown in Appendix H. So, the collected data was met requirement of this
assumption.
Assumption 3:
Multicollinearity happens when two or more predictors contain much of the
same information. In other word, there are the high intercorrelations among some
composite of the independent variables. According to the correlation matrix below,
correlations among predictors were low and Pearson test indicated the value lower
than 0.7. This means that there is a low possibility with multicollinearity.
Table 4.6:
Correlations Matrix
CV
CV
IV
1
IV
ICP
.444
.467
1
.663
ICP
(IV: Individualistic Values
1
CV: Collectivistic Values
ICP: Individual consequences of the purchase)
In other hand, assumption about multicollinearity is also tested by evaluating
two collinearity diagnostic factors: Tolerance and the Variance Inflation Factor –
VIF). The Variance Inflation Factor (VIF) performs the influence of collinearity
among the variables in a regression model. When Tolerance value is less than 0.1 or
VIF higher than 10, there is high possibility of mutlicollinearity. If VIF less than 4,
there is no multicollinearity. In this research, as table 4.7, the result showed that
29
PERSONAL VALUEATTITUDEENVIRONMENTAL COMMITMENT
30
tolerance value was 0.803 and VIF was 1.245. As a result, there was not
multicollinear. Thus, this assumption is satisfied.
Table 4.7:
Coefficients Matrix
Unstandardized
Coefficients
Model
1
B
Standardized
Coefficients
Std.Error
Beta
Collinearity
Statistics
T
Sig. Tolerance
VIF
(Constant)
2.560
.518
4.943 .000
CV
.321
.099
.215
3.235 .002
.803
1.245
IV
.583
.068
.568
8.529 .000
.803
1.245
a. Dependent Variable: ICP
(IV: Individualistic Values
CV: Collectivistic Values
ICP: Individual consequences of the purchase)
In brief, collected data was satisfied with 3 main assumptions to run multiple
regression test.
After checking all assumptions, the result of running multiple regression was
reported to determine how well the model fit.
Table 4.8:
Model Summary
Model
1
R
R Square
.691a
a. Predictors: (Constant), CV, IV
b. Dependent Variable: ICP
(IV: Individualistic Values
.477
Adjusted R
Square
.470
Std. Error of
the Estimate
.79664
DurbinWatson
2.063
PERSONAL VALUEATTITUDEENVIRONMENTAL COMMITMENT
31
CV: Collectivistic Values
ICP: Individual consequences of the purchase)
According to the Model Summary table, the multiple correlation coefficient (R)
was 0.691, R Square was 0.477, and adjusted R Squared was 0.470. R Square was
0.477; this is showing that 47.7% of the variance in individual consequences of
purchase could be predicted from two independent variables: collectivistic values and
individualistic values.
Table 4.9:
ANOVA
Sum of
Squares
Model
1
Mean
Square
df
Regression
85.079
2
42.539
Residual
93.291
147
.653
178.370
149
Total
F
67.030
Sig.
.000b
a. Dependent Variable: ICP
b. Predictors: (Constant), CV, IV
(IV: Individualistic Values
CV: Collectivistic Values
ICP: Individual consequences of the purchase)
The value of F was 67.030 and sig. was 0.000 (less than 0.05). This indicates
the combination of these variables significantly predicts the dependent variable.
H3: Individualistic values will relate positively to attitude toward the individual
consequences of the purchase.
According to Coefficients matrix (Table 4.7), the factor of individualistic
values had sig value around 0.000 which was less than 0.05 as requirement. This
meant individualistic values impact on attitude toward the individual consequences.
PERSONAL VALUEATTITUDEENVIRONMENTAL COMMITMENT
And the results showed that individualistic values were positively related to attitude
toward the individual consequences of the purchase because its Standardized
Coefficient was positive (beta was 0.568). This implied that this factor which
investigated in this research was a meaningful factor to consumers’ attitude toward the
individualistic consequences of the purchase. Therefore, the hypothesis 3 was well
confirmed.
H2: Collectivistic values will relate negatively to attitude toward the individual
consequences of the purchase.
In order to evaluate the impact of collectivistic values, the result showed that
the sig value of this factor was significant (0.002 which less than 0.05). This meant
collectivistic values impact on attitude toward the individual consequences. Moreover,
this relationship was negative because beta was 0.215. This finding was the same as
hypothesis. Thus, the hypothesis 2 was supported.
4.4.2 Multiple regression analysis: to evaluate the impact of collectivistic values
and individualistic values on the attitude toward the environmental consequences
of the purchase.
Results of testing assumptions
Assumption 1:
This assumption could be checked by using Curve Estimation Graphs and
Scatter Graphs to define the relationship between the environmental consequences of
the purchase and each predictor. As the plot shown in Appendix I, the relationship
between the environmental consequences of the purchase and each predictor was
linear regression. Therefore, this assumption was satisfied.
32
PERSONAL VALUEATTITUDEENVIRONMENTAL COMMITMENT
Assumption 2:
The normal distribution of dependent variable was test by histogram and scatter
plot, which shown in Appendix I. The result of this test was support for the normal
distribution.
Assumption 3:
It is obligated to test correlation between variables by using Pearson
Correlation. The result showed that explanatory variables are not corrected with each
other.
Table 4.10:
Correlations Matrix
CV
CV
IV
1
IV
ECP
.444
.671
1
.475
ECP
1
(IV: Individualistic Values
CV: Collectivistic Values
ECP: Environmental consequences of the purchase)
Moreover, assumption about multicollinearity is also tested by evaluating
Tolerance and the Variance Inflation Factor – VIF). In this research, as table 4.11, the
result showed that tolerance value was 0.803 and VIF was 1.245. As a result,
multicollinear really dit not happen in the collected data. Thus, this assumption is
satisfied.
33
PERSONAL VALUEATTITUDEENVIRONMENTAL COMMITMENT
34
Table 4.11:
Coefficients Matrix
Unstandardized
Coefficients
Model
1
(Constant)
B
Standardized
Coefficients
Std.Error
Beta
Collinearity
Statistics
T
Sig. Tolerance
VIF
1.465
.404
3.627 .000
CV
.674
.077
.573
8.719 .000
.803
1.245
IV
.179
.053
.221
3.358 .001
.803
1.245
b. Dependent Variable: ECP
(IV: Individualistic Values
CV: Collectivistic Values
ECP: Environmental consequences of the purchase)
In brief, collected data was satisfied with 3 main assumptions to run multiple
regression test.
Next, the result of running multiple regression was reported to determine how
well the model fit.
Table 4.12:
Model Summary
Model
1
R
R Square
.700a
.490
Adjusted R
Square
.483
c. Predictors: (Constant), CV, IV
d. Dependent Variable: ECP
(IV: Individualistic Values
CV: Collectivistic Values
ECP: Environmental consequences of the purchase)
Std. Error of
the Estimate
.62137
DurbinWatson
1.909