CHAPTER 6: DATA ANALYSIS AND FINDINGS

1. Data analysis and findings

Kaiser-Meyer-Olkin (KMO) test (Table 6.1.1) is a measure of the sample adequacy for
performing factor analysis. The test measures sample adequacy for each variable in the model
and for the complete model. KMO return values between 0 to 1. The rule of thumb for
interpreting the result is KMO values in the range of 0.8 to 1 is indicate that the sample is
adequate.

Table 6.1.1 KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .840

Approx. Chi-Square 3253.556

Bartlett's Test of Sphericity Df 276
Sig. .000

The Bartlett's Test is used to test whether the sample have equal variances. Equal variance
across sample is called homogeneity of variances. Since the Bartlett's Test shows significant P
value (0.000) the variance are not equal.

Table 6.1.2 Communalities

Initial Extractio
n
OF1 1.000 .500
OF2 1.000 .662
OF3 1.000 .515
TF1 1.000 .793
TF2 1.000 .815
TF3 1.000 .765
SF1 1.000 .675
SF2 1.000 .622
SF3 1.000 .711
TR1 1.000 .691
TR2 1.000 .790
TR3 1.000 .699
TP1 1.000 .739
TP2 1.000 .653
TP3 1.000 .807
CP1 1.000 .614
CP2 1.000 .744
CP3 1.000 .670
AP1 1.000 .692
AP2 1.000 .796
AP3 1.000 .602
+CPW
1.000 .815
B1
CPWB
1.000 .856
2
CPWB
1.000 .801
3
Extraction Method: Principal
Component Analysis.

In (Table 6.1.2) communality for a given variable can be interpreted as the proportion of
variation in that variable explained by the eight factors. It can be obtained from this
communality that how accurately this model performs. If the communalities values for
individual variables are close to 1, indicates that the variables can be included in the model
explained the total variance. Since the communalities values are more than 0.5 indicates that
the variables explaining the eight factors are good.
Table 6.1.3 Total Variance Explained
Compone Initial Eigenvalues Extraction Sums of Rotation Sums of
nt Squared Loadings Squared Loadings
Tota % of Cumulati Tota % of Cumulati Tota % of Cumulati
l Varian ve % l Varian ve % l Varian ve %
ce ce ce
6.47 6.47 2.62
1 26.983 26.983 26.983 26.983 10.934 10.934
6 6 4
2.23 2.23 2.57
2 9.307 36.291 9.307 36.291 10.721 21.655
4 4 3
1.97 1.97 2.18
3 8.223 44.514 8.223 44.514 9.110 30.765
4 4 6
1.67 1.67 2.09
4 6.986 51.500 6.986 51.500 8.715 39.480
7 7 2
1.53 1.53 2.03
5 6.383 57.883 6.383 57.883 8.484 47.964
2 2 6
1.20 1.20 1.89
6 5.005 62.888 5.005 62.888 7.884 55.848
1 1 2
1.08 1.08 1.87
7 4.516 67.404 4.516 67.404 7.808 63.657
4 4 4
1.74
8 .850 3.541 70.945 .850 3.541 70.945 7.288 70.945
9
9 .800 3.333 74.277
10 .630 2.627 76.904
11 .609 2.536 79.440
12 .592 2.467 81.907
13 .544 2.266 84.173
14 .519 2.163 86.336
15 .486 2.025 88.362
16 .463 1.928 90.290
17 .395 1.647 91.937
18 .379 1.578 93.516
19 .326 1.357 94.872
20 .297 1.238 96.111
21 .275 1.144 97.255
22 .250 1.041 98.296
23 .223 .930 99.226
24 .186 .774 100.000
Extraction Method: Principal Component Analysis.

The above figure (Table 6.1.3) interpret that total variance explained by eight factors are
70.94%. As for as Eigen value, there are only seven factor loaded but these extracted eight
factors based on the literature support (Glyptis et al., (2020); Putri and Idulfilastri (2020);
Gupta et al., (2017); Acosta and Torres (2017); Kamal et al., (2013); Al-Shafi, and Weerakkody
(2010); AlAwadhi and Morris (2008). These eight factors explained the total model about 71%.

Table 6.1.4 Rotated Component Matrixa

Component
1 2 3 4 5 6 7 8
OF1 .605

I OF2 .737

OF3 .622
TF1 .839

II TF2 .842
TF3 .835
SF1 .774
III SF2 .633
SF3 .796
TR1 .760
IV TR2 .845
TR3 .727
TP1 .726
V TP2 .705
TP3 .850
CP1 .681
VI CP2 .806
CP3 .796
AP1 .719
VII AP2 .869
AP3 .554
CPWB1 .870

CPWB2 .892
VIII
CPWB3 .863
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
a. Rotation converged in 6 iterations.

In (Table 6.1.4) rotated component matrix showed that there are eight factors loaded and three
items are explaining each factors. The factors are not correlated with other factors; this shows
that the independency. The first factor is correlated with three items such as TF1, TF2 and TF3.
The second factors explained by the items CPWB1, CPWB2 and CPWB3. The third factor is
correlated with the items TR1, TR2 and TR3. The fourth factors is correlated with the following
items TP1, TP2 and TP3. The fifth factor is explained by the items CP1, CP2 and CP3. The
variables SF1, SF2 and SF3 are highly correlated with the sixth factor. The variables OF1, OF2
and OF3 are correlated with seventh factor and the eight factor explained by the following
variables AP1, AP2 and AP3.

According to the literature review, it is observed that factors 1, 3, 6 and 7 explained the latent
factor EGA, and the factors 2, 4, 5 and 8 are predicting the latent factor EJP.

Internal consistency assessment

After conducing EFA, the internal consistency of all the extracted factors was assessed using
Cronbach’s alpha coefficient. As per the Tables 6.1.5, the rule of thumb that values of all
Cronbach’s alpha coefficients exceed the threshold value of 0.70 indicating adequate internal
consistency for all the factors.
Table 6.1.5: Internal consistency assessment

Factor No. of Items Cronbach's Alpha

OF 3 0.705
TF 3 0.875
SF 3 0.718
TR 3 0.788
TP 3 0.781
CP 3 0.727
AP 3 0.894
CPWB 3 0.703

Discriminant Validity: Fornell-Larcker Criterion

Factor AP CP CPWB OF SF TF TP TR
AP 0.789
CP 0.335 0.804
CPWB 0.276 0.292 0.908
OF 0.323 0.157 0.174 0.748
SF 0.162 0.290 0.242 0.478 0.799
TF 0.450 0.241 0.211 0.421 0.237 0.894
TP 0.354 0.310 0.369 0.329 0.187 0.294 0.835
TR 0.364 0.268 0.297 0.333 0.225 0.370 0.488 0.839

Confirmatory factor analysis results for EGA and JPD

Fitness index Recommended criterion Obtained values

chi square 343.182
Df 98
chi square/df 2-5 3.502
GFI 0.900 0.915
AGFI 0.900 0.882
NFI >=0.95 0.955
TLI >= 0.95 0.935
CFI >=0.95 0.948
RMR 0.042
RMSEA 0.079
Convergent and Discriminant validity:

 Construct validity = Is the extent to which a set of measured variables actually represent
the theoretical latent construct they are designed to measure. It is made up of four
components: convergent validity, discriminant validity, nomological validity and face
validity.
 Convergent validity = The extent to which indicators of a specific construct ‘converge’ or
share a high proportion of variance in common.
 Face validity = The extent to which the content of the items is consistent with the construct
definition, based solely on the researcher’s judgment.
 Average Variance Extracted (AVE) = a summary measure of convergence among a set
of items representing a construct. It is the average percent of variation explained among
the items.
One of the biggest advantages of CFA/SEM is its ability to quantitatively assess the construct
validity of a proposed measurement theory. Construct validity is the extent to which a set of
measured items actually reflect the theoretical latent construct they are designed to measure.

The AVE (Average Variance Explained) is not provided by AMOS software, so it has to be
calculated. The AVE is calculated as the mean variance extracted for the items loading on a
construct and is a summary indicator of convergence (Fornell and Larcker, 1981).

A good rule of thumb is an AVE of .5 or higher indicates adequate convergent validity. An

AVE of less than .5 indicates that on average, there is more error remaining in the items than
there is variance explained by the latent factor structure have imposed on the measure. An AVE
estimate should be computed for each latent construct in a measurement model.
Table 6.1.6 Confirmatory Factor Analysis:
Table 6.1.7 Regression Weights: (Group number 1 - Default model)

Estimate S.E. C.R. P Label

OF3 <--- OF 1.000
OF2 <--- OF 1.177 .153 7.692 *** par_1
OF1 <--- OF 1.078 .149 7.213 *** par_2
TF3 <--- TF 1.000
TF2 <--- TF 1.059 .063 16.686 *** par_3
TF1 <--- TF 1.039 .064 16.256 *** par_4
SF3 <--- SF 1.000
SF2 <--- SF .919 .096 9.548 *** par_5
SF1 <--- SF .910 .101 8.970 *** par_6
TR3 <--- TR 1.000
TR2 <--- TR 1.227 .100 12.232 *** par_7
TR1 <--- TR 1.093 .097 11.212 *** par_8
TP3 <--- TP 1.000
TP2 <--- TP .664 .064 10.323 *** par_9
TP1 <--- TP 1.064 .081 13.137 *** par_10
CP3 <--- CP 1.000
CP2 <--- CP .999 .111 9.010 *** par_11
CP1 <--- CP .879 .101 8.696 *** par_12
CPWB3 <--- CPWB 1.000
CPWB2 <--- CPWB 1.102 .058 18.946 *** par_13
CPWB1 <--- CPWB 1.054 .060 17.605 *** par_14
AP3 <--- AP 1.000
AP2 <--- AP .830 .103 8.034 *** par_15
AP1 <--- AP .970 .103 9.461 *** par_16
Table 6.1.8 Standardized Regression Weights: (Group number 1 - Default model)

Estimate
OF3 <--- OF .538
OF2 <--- OF .651
OF1 <--- OF .577
TF3 <--- TF .801
TF2 <--- TF .867
TF1 <--- TF .840
SF3 <--- SF .698
SF2 <--- SF .703
SF1 <--- SF .630
TR3 <--- TR .732
TR2 <--- TR .815
TR1 <--- TR .702
TP3 <--- TP .781
TP2 <--- TP .607
TP1 <--- TP .837
CP3 <--- CP .608
CP2 <--- CP .793
CP1 <--- CP .667
CPWB3 <--- CPWB .819
CPWB2 <--- CPWB .919
CPWB1 <--- CPWB .841
AP3 <--- AP .802
AP2 <--- AP .514
AP1 <--- AP .624
 Table 6.1.9 Model Summary : Confirmatory Factor Analysis

Acceptable
Output of
S.No Measures of fit level for good
EMPENG Model
fit
1 Chi-square ( χ2) at p 0.01 343.182
2 Degree of freedom (d.f) 98
3 Chi-square/ Degree of freedom (d.f) 3.502 2-5
4 Goodness of Fit Index (GFI) 0.915 0.900
5 Adjusted Goodness of Fit Index (AGFI) 0.882 0.900
6 Comparative Fit Index (CFI) 0.948 >=0.95
Bentler – Bonett Index or Normed Fit
7 0.955
Index (NFI) >=0.95

8 Tucker-Lewis Index (TLI) 0.935 >=0.95

9 Root Mean Square Residual (RMR) 0.042

Root mean squared error of

10
approximation (RMSEA) 0.079
Source: AMOS 21.0 Output
As per table (Table 6.1.10) Average Variance Extracted (AVE) estimates all exceed .5 or close
to 0.5. In addition, the model fits relatively well based on the Goodness of Fit measure.
Therefore, all the indicator items are retained at this point and adequate evidence of convergent
validity is provided. As such further modeling and analysis can be proceeded with confidence
that the questionnaire measures these key constructs well.
 Table 6.1.10 Convergent and Discriminant validity
Indicator Latent Standardized AVE CR Cronbach's
Variables Variables Loadings Alpha
OF3 <--- OF 0.701
OF2 <--- OF 0.796 0.560 0.792 0.705
OF1 <--- OF 0.745
TF3 <--- TF 0.881
TF2 <--- TF 0.903 0.800 0.923 0.875
TF1 <--- TF 0.899
SF3 <--- SF 0.817
SF2 <--- SF 0.806 0.639 0.841 0.718
SF1 <--- SF 0.775
TR3 <--- TR 0.822
TR2 <--- TR 0.880 0.704 0.877 0.788
TR1 <--- TR 0.813
TP3 <--- TP 0.867
TP2 <--- TP 0.774 0.697 0.873 0.781
TP1 <--- TP 0.860
CP3 <--- CP 0.746
CP2 <--- CP 0.848 0.647 0.845 0.727
CP1 <--- CP 0.815
CPWB3 <--- CPWB 0.893
CPWB2 <--- CPWB 0.931 0.825 0.934 0.894
CPWB1 <--- CPWB 0.900
AP3 <--- AP 0.830
AP2 <--- AP 0.713 0.623 0.832 0.703
AP1 <--- AP 0.819
Discriminant validity:

To test the discriminant validity, we need square root of Average Variance Extracted (AVE)
and then will compare with the correlation between the latent construct. A good rule of thumb
a square root of AVE is higher than the correlation value between the latent construct.

Table 6.1.11 : Discriminant validity

Latent Latent Correlation

<-->
Variables Variables Estimates
OF <--> CPWB 0.248
OF <--> AP 0.518
OF <--> CP 0.23
OF <--> TP 0.474
OF <--> TR 0.457
OF <--> SF 0.724
OF <--> TF 0.555
TF <--> CPWB 0.24
TF <--> AP 0.599
TF <--> CP 0.301
TF <--> TP 0.328
TF <--> TR 0.436
TF <--> SF 0.291
SF <--> CPWB 0.29
SF <--> AP 0.248
SF <--> CP 0.377
SF <--> TP 0.245
SF <--> TR 0.296
TR <--> CPWB 0.34
TR <--> AP 0.502
TR <--> CP 0.367
TR <--> TP 0.576
TP <--> CPWB 0.435
TP <--> AP 0.445
 TP <--> CP 0.371
CP <--> CPWB 0.329
CP <--> AP 0.465
CPWB <--> AP 0.354

Table 6.1.12 Square root of Average Variance Extracted (AVE)

Factors AP CP CPWB OF SF TF TP TR
AP 0.789
CP 0.335 0.804
CPWB 0.276 0.292 0.908
OF 0.323 0.157 0.174 0.748
SF 0.162 0.290 0.242 0.478 0.799
TF 0.450 0.241 0.211 0.421 0.237 0.894
TP 0.354 0.310 0.369 0.329 0.187 0.294 0.835
TR 0.364 0.268 0.297 0.333 0.225 0.370 0.488 0.839

The above tables show that the square root of AVE is higher than the correlation between the
construct. So it is interpreted the discriminant validity for the latent construct is exist.

Structural equation model

Hair et al., (2016) stated the role of the structural equation model in determining how well the
data collected supports the theory. Several researchers have used to structural equation model
because of its unique characteristics and advantages. Structural equation model provides a
direct approach to manage relationships simultaneously; hence it is able to provide statistical
efficiency concurrently which is not applicable in multiple regression analysis. It also examines
comprehensively the relationships between the observed and latent variables (Hoyle, 1995;
Schaupp et al., 2010). Moreover, it allows an easy shift from exploratory factor analysis to
confirmatory factor analysis. Structural equation model also demonstrates the concepts that are
not observed through these associations and justify the measurement error in the estimation
process which are not achievable in the multiple regression analysis (Kline, 2001; Prajogo and
Cooper, 2010), SEM offers comprehensive information about the research model that extends
beyond the regression method (Bollen, 1989; Jimenez-Jimenez and Martinez-Costa, 2009;
Jöreskog and Sörbom, 1993)

Evaluation of model fit

Model fit indicators depict various indicators of goodness of fit for the proposed model along
with their acceptance values. In order to evaluate model fit, various well-known goodness-of-
fit indices are used (Byrne, 1998; Hair et al., 1992; Jöreskog and Sörbom, 1996). These include
the chi-square χ2, the comparative fit index, the unadjusted goodness-of-fit indices, the normal
fit index, the Tucker-Lewis Index, the root mean square error of approximation and the
standardized root mean square error residual. Goodness-of-fit indices provide cutoff values to
evaluate data-model fit. Hu and Bentler, 1999 recommended using combinations of Goodness-
of-fit indices to obtain a comprehensive evaluation of model fit. The criterion values for with
good model fit for in structural equation modeling are comparative fit index (CFI) > 0.95,
Tucker-Lewis Index (TLI) > 0.95, root mean square error of approximation (RMSEA) < 0.06,
and Square Error Residual (SRMR) < 0.08 . Some researchers suggest that these cutoff values
are too stringent and that the findings may have limited generalizability (Beauducel and
Wittmann, 2005; Fan and Sivo, 2005; Marsh et al., 2004). Therefore, it is recommended that
absolute fit indices and incremental fit indices (such as CFI, GFI, NFI, and TLI), the cut off
values should be above 0.90 and fit indices based on residuals matrix values (such as RMSEA
and SRMR) should be below 0.10 or 0.05. Figure 2 depicts the model with standardized
regression weights, which was extracted from AMOS graphics.
Figure 6.1.1 Structural Equation Modeling

The path diagram shows the impact of e-government adoption on job performance among the
employees of e-district in Uttarakhand. The dimensions of E-Government Adoption (EGA) are
Organizational Factor (OF); Technical Factor (TF); Trust Factor (TR) and Social Factor (SF).
The dimensions of Job Performance (JP) include Task performance (TP), Contextual
Performance (CP), Adaptive Performance (AP) and Counter-Productive Work Behaviour
(CPWB). The Root Mean Square Error of Approximation (RMSEA) fit statistics for the model
was 0.076, which is considered as the best fit model (Brown and Cudeck, 1993;
Diamantopoulos and Siguaw, 2000). The path diagram shows the impact of e-government
adoption is 10.779 per cent on job performance among the employees of e-district in
Uttarakhand. It can also be seen that E-government adoption (EGA) is influenced by all factors
(organizational factor, technical factor, trust factor and social factor). The regression weights,
standardized regression weights and the model fit summary are shown in Table 6.1.11, 6.1.12
and 6.1.13 respectively. It is concluded from the Table 11, 12 and 13, all the statistical measures
are in the acceptable range for the structural equation model.

Path coefficients for structural model

Relationship Path Coefficient Critical Result

(Standardized) Ratio
OF EGA 0.757 21.756*** H1 Accepted – Significant

TF EGA 0.763 23.320*** H1 Accepted – Significant

TR EGA 0.690 15.785*** H1 Accepted – Significant

SF EGA 0.631 12.298*** H1 Accepted – Significant

EGA JPD 0.566 10.799*** H1 Accepted – Significant

TP JPD 0.741 19.876*** H1 Accepted – Significant

CP JPD 0.656 13.006*** H1 Accepted – Significant

AP JPD 0.670 14.295*** H1 Accepted – Significant

CPWB JPD 0.731 16.157*** H1 Accepted – Significant

Note: ***p<0.001; **p<0.01

Table 6.1.13 Regression Weights: (Group number 1 - Default model)

Relationship Path Coefficient Critical Result

(Standardized) Ratio

OF EGA 0.757 21.756*** H1 Accepted – Significant

TF EGA 0.763 23.320*** H1 Accepted – Significant

TR EGA 0.690 15.785*** H1 Accepted – Significant

SF EGA 0.631 12.298*** H1 Accepted – Significant

EGA TP 0.419 19.876*** H1 Accepted – Significant

EGA CP 0.371 13.006*** H1 Accepted – Significant

EGA AP 0.379 14.295*** H1 Accepted – Significant

EGA 0.413 16.157*** H1 Accepted – Significant

CPWB

Note: ***p<0.001; **p<0.01

Table 6.1.14 Standardized Regression Weights: (Group number 1 - Default model)

Estimate
JPD <--- EGA .840
OF <--- EGA 1.000
TF <--- EGA 1.000
SF <--- EGA 1.000
TR <--- EGA 1.000
TP <--- JPD 1.000
CP <--- JPD 1.000
 Estimate
AP <--- JPD 1.000
CPWB <--- JPD 1.000
OF3 <--- OF .569
OF2 <--- OF .544
OF1 <--- OF .622
TF3 <--- TF .650
TF2 <--- TF .663

TF1 <--- TF .539

SF3 <--- SF .520
SF2 <--- SF .662
SF1 <--- SF .619
TR3 <--- TR .593
TR2 <--- TR .511
TR1 <--- TR .573
TP1 <--- TP .600
TP2 <--- TP .555

TP3 <--- TP .630

CP1 <--- CP .597
CP2 <--- CP .593
CP3 <--- CP .572
AP1 <--- AP .528
AP2 <--- AP .608
AP3 <--- AP .596

CPWB1 <--- CPWB .582

CPWB2 <--- CPWB .625
CPWB3 <--- CPWB .663
Table 6.1.15 Model Summary : Structure Equation Modelling (SEM)
S.No Measures of fit Output of Acceptable level
EMPENG for good fit
Model
1 Chi-square ( χ2) at p 0.01 940.942
2 Degree of freedom (d.f) 251
3 Chi-square/ Degree of freedom (d.f) 3.75 2-5
4 Comparative fit index (CFI) 0.953 >=0.95
5 Goodness of Fit Index (GFI) 0.941 0.900
6 Adjusted Goodness of Fit Index (AGFI) 0.923 0.900
7 Tucker-Lewis Index (TLI) 0.952 >=0.95
8 Root Mean Square Residual (RMR) 0.036 0.050
9 Root mean squared error of 0.057 0.080
approximation (RMSEA)
Source: AMOS 21.0 Output : Structural Equation Modeling: Model Fit

Fit Index Threshold value Obtained value

chi square 940.942
Df 251
chi square/df 2-5 3.74
GFI 0.900 0.941
AGFI 0.900 0.923
TLI >= 0.95 0.952
CFI >=0.95 0.953
RMR 0.050 0.036
RMSEA 0.080 0.057
Figure 6.1.2 Regression Path Analysis

In Figure 6.1.2 regression path analysis shows that the Organizational factor (OF) is major
influencing factor on the Employee’s Job performance (EJP) as the regression weight assigned
0.405, the next social factors (SF) is influencing the predictor variable Employee job
performance (EJP) 0.341 followed by Trust factors (0.323) and Technical factors (0.306)
influence the dependent factors employees job performance (EJP). The visual representations
of results suggest that the relationships between the constructs of adoption of E-governance
services are stronger. The overall regression model shows that the e-governance services
factors organizational factor (OF), social factor (SF), trust factor (TR) and technical factors
(TF) significantly influence the dependent factor employee job performance (EJP) and model
represented R2 0.960 i.e. 96%, the independent factors are explaining the dependent factor
(EJP).

Table 6.1.16 Regression Weights: (Group number 1 - Default model)

Estimate S.E. C.R. P Label

EJP <--- OF 1.397 .042 33.450 ***
 Estimate S.E. C.R. P Label
EJP <--- TF 1.166 .048 24.496 ***
EJP <--- SF 1.300 .048 27.111 ***
EJP <--- TR 1.422 .060 23.858 ***

Table 6.1.17 Standardized Regression Weights: (Group number 1 - Default model)

Estimate
EJP <--- OF .405
EJP <--- TF .306
EJP <--- SF .341
EJP <--- TR .323

Table 6.1.18 Squared Multiple Correlations: R2: (Group number 1 - Default model)

Estimate
EJP .960

Interpretation:

In hierarchical regression, the predictor variables are entered in sets of variables according to
a pre-determined order that may infer some causal or potential relationships between the
predictors and the dependent variable (Francis, 2003). Such situations are area of interest in the
social sciences. Hence, the researcher empirically tested the hierarchical regression for the
model conceptualized in the regression path analysis within the AMOS graphics environment.
The analyses conducted, the parameter estimates are then viewed within AMOS graphics and
it displays the standardized parameter estimates. The regression analysis revealed that the
impact of the factors influencing adoption of E-Governance services on the employee’s job
performance.
Chapter summary 6

OBJECTIVE OF HYPOTHESIS STATISTICAL RESULT

THE STUDY TOOL (Status of
hypothesis testing)
1) To e-Government H1a: Organizational
Adoption Factor is significantly
important for e- SEM ACCEPTED
government adoption
among employees.

H2b: Technical Factor

is significantly
important for e- SEM ACCEPTED
government adoption
among employees.

H3c: Social Factor is

significantly important
for e-government
adoption among SEM
employees. ACCEPTED

H4d: Trust Factor is

significantly important
for e-government ACCEPTED
adoption among SEM
employees.
2) To assess the job H2a: Task
performance of performance is
employees significantly important SEM ACCEPTED
for the employee’s job
performance.

H2b: Contextual
performance is
significantly important ACCEPTED
for employee's job SEM
performance.
H2c: Adaptive
behavior is
significantly important ACCEPTED
for employee’s job SEM
performance.

H2d: Counter-
productive behavior is
significantly important ACCEPTED
for employee’s job SEM
performance.