You are on page 1of 9

SHORT TEST 2

Exploratory Factor Analysis.........................................................................................................................1


Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO)...................................................................1
Convergent validity..................................................................................................................................1
Discriminant Validity................................................................................................................................1
Predictive validity....................................................................................................................................2
REGRESSION ANALYSIS................................................................................................................................3
Regression Assumptions..........................................................................................................................3
Multiple Linear Regression Analysis............................................................................................................4
Innovation Performance (IP)....................................................................................................................4
Supply chain performance.......................................................................................................................5
HYPOTHESIS.................................................................................................................................................6

1
QUESTION#1
Exploratory Factor Analysis

Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO)


Factor Analysis of all dependent and independent variables resulted in KMO’s measure of
sampling adequacy to be 0.820, which indicates the sample is suitable for factor analysis.
Bartlett’s test of Sphericity is significant p (0.000) < α, which indicates factor analysis can be
performed for this data.

Convergent validity:
Table 1 shows that retained variables of each construct are significantly related (highest loading
in desired scales) to each other, which means convergent validity is present.

Discriminant Validity:
SOC1, SOC6, SC4, SC5, SC6, SI1, RG1, RG2, RG3.are eliminated which had significant
loading or non desired constructs. Some values of these variables were different from other
constructs which we removed, so discriminant validity is also present between variables.
Table 1
Rotated Component Matrix
Supply
chain
performanc Social Supply chain Innovation Supplier
Items   e capital strategy performance integration
SC1       0.809    
SC2       0.81    
SC3       0.697    
SOC2     0.576      
SOC3     0.643      
SOC4     0.799      
SOC5     0.766      
SI2           0.778

SI3           0.863

IP1         0.809  
IP2         0.86  
SCP1  
0.758        
SCP2  
0.825        
SCP3  
0.88        
SCP4  
0.721        
SCP5  
0.652        

2
SCP64  
0.724        

Predictive validity:
1. Innovation performance:
Table#2 describes that there is significant relationship between innovation performance and
independent variables (SC, SOC, and SI).

Table 2

Correlation
      s        
  Mean Std. deviation SC SOC SI IP
1.0072
1
SC 5.6183 1
SOC 5.9418 0.807 .483** 1
1.3512
.182* .179* 1
SI 4.9113 5
1.6177
.142 .228** .181* 1
IP 4.7198 7

2. Supply chain performance

Table#3 provides a correlation matrix and descriptive statistics for Supply Chain Performance
(SCP) so there is significant relationship between SCP and independent variables (SC, SOC, and
SI).

Table 3

Correlations
Std.
Deviatio
n Mean SC SOC SI SCP
SC 5.62 1.01 1
SOC 5.94 0.81 .483** 1
SI 4.91 1.35 .182 *
.179* 1

3
SCP 5.05 1.08 .072 .256** .130 1

QUESTION#2

REGRESSION ANALYSIS

Regression Assumptions:
i. Homoscedasticity: refers to whether these residuals are equally distributed, or whether

they tend to bunch together at some values, and at other values, spread far apart. Our data
is homoscedastic as it looks randomly distributed, it does not have an obvious pattern,
there are points equally distributed on X-axis and Y-axis.
ii. Normality of data: To check the normality of data we plotted a histogram which shows
a well bell shaped pattern and data is normally distributed which is little leftward skewed.
iii. Linearity: means that the predictor variables in the regression have a straight-line
relationship with the outcome variable. Our residuals are normally distributed and
homoscedastic.
iv. Multicollinearity: refers to weather there is correlation between Independent variables
and dependent variables we check this assumption through VIF as our VIF values are
below 10 so there is no multicollinairity between variables.

4
Multiple Linear Regression Analysis:
Innovation Performance (IP)
In this Table, The overall strength of the relationship between the predictors and the criterion is
10.9 % (R square = 0.109), indicating weak relationship. All the independent variables together
explained only 10.3% of variation in dependent variable. From goodness of fit view point this
strength is better measured at 8.9% (Adjusted R Square = 0.089), this also indicates weak
association of the relationship between the predictors and the criterion variables.
Following table shows that SC has insignificant values while SOC, RG and SI have significant
value of .513, -.308 and .318 respectively because (P< .005)

Table
4
Results of regression analysis Model 1
  Components Unstandradized β Standarized β t-value Sign.  

(Constant) 1.207 1.278 0.203


SC 0.083 0.052 0.606 0.546
SOC 0.513 0.265 3 0.003
RG -0.308 -0.182 -2.014 0.046
SI 0.318 0.207 2.72 0.007
Dependent variable: IP
R² .109
Adjusted R² .089
F-value, Sig
0.000 5.414

5
Supply chain performance:
In this Table, The overall strength of the relationship between the predictors and the criterion is
10.3 % (R square = 0.103), indicating weak relationship. All the independent variables together
explained only 10.3% of variation in dependent variable. From goodness of fit view point this
strength is better measured at 8.3% (Adjusted R Square = 0.083), this also indicates weak
association of the relationship between the predictors and the criterion variables.
Following table 4 shows that SC, RG and SI have insignificant values while SOC have
significant value of .322 because (P< .005)

Table
5
Results of regression analysis model 2
  Components Unstandradized β Standarized β t-value Sign.

(Constant) 2.636 4.178 0


SC -0.139 -0.130 -1.515 0.131
SOC 0.322 0.250 2.814 0.005
RG 0.127 0.112 1.239 0.217
SI 0.111 0.109 1.428 0.155

R² 0.103
Adjusted R² 0.083
F-value, Sig
0.001 5.078

6
HYPOTHESIS:
Tables summarize the hypothesized relationship, regression estimate and t value for model 1 and
model 2,

H1 of Model-1

We hypothesized that Rational Governance has positive impact on Innovation Performance. Our
results indicate that RG is contribute negatively toward IP (β= -2.014; P< 0.05), so significant
relationship between these variables. Overall, H1 is being fully supported.

H2 of Model-1,

We hypothesized that Supply chain Strategy has positive impact on Innovation Performance.
There is absence of the significant relationship between the SC & IP (β= 0.606; P> 0.05),
Overall, H2 is being not supported.

H3 of Model-1,

We hypothesized that Social Capital has positive impact on Innovation Performance. Our results
indicate that SOC is contributing positively toward IP (β= 3.00; P< 0.05), so significant
relationship between these variables. Overall, H3 is being fully supported.

H4 of Model-1,

We hypothesized that Supplier Integration has positive impact on Innovation Performance. Our
results indicate that SI is contributing positively toward IP (β= 2.720; P< 0.05), so significant
relationship between these variables. Overall, H4 is being fully supported.

H1 of Model-2,

7
We hypothesized that Rational Governance has positive impact on Supply Chain Performance.
There is absence of the significant relationship between the RG & SCP (β= 0.112; P> 0.05),
Overall, H1 is being not supported.

H2 of Model-2,

We hypothesized that Supply Chain Strategy has positive impact on Supply Chain Performance.
There is absence of the significant relationship between the SC & SCP (β= -0.130; P> 0.05),
Overall, H2 is being not supported.

H3 of Model-2

We hypothesized that Social Capital has positive impact on Supply Chain Performance. Our
results indicate that SOC is contributing positively toward IP (β= 0.250; P< 0.05), so significant
relationship between these variables. Overall, H3 is being fully supported.

H4 of Model-2

We hypothesized that Supplier Integration has positive impact on Supply Chain Performance.
There is absence of the significant relationship between the SI & SCP (β= 0.109; P> 0.05),
Overall, H4 is being not supported.

Table 6
Standardized
Models   Hypothesis β   t-value Sign
IP H1 RG→IP -2.014 -2.014 .046
H2 SC→IP .606 .606 .546
H3 SOC→IP 3.000 3.000 .003
H4 SI→IP 2.720 2.720 .007

SCP H1 RG→SCP .112 1.239 .217


H2 SC→SCP -.130 -1.515 .131
H3 SOC→SCP .250 2.814 .005
H4 SI→SCP .109 1.428 .155

8
9

You might also like