Types of uncertainty

Code

Component


1

2

3

4

5

6

7


Climate uncertainty (CLU)

CLU3

0.890
      
CLU2

0.889
      
CLU4

0.887
      
CLU1

0.830
      
Planning and control uncertainty (PCU)

PCU2
 
0.953
     
PCU3
 
0.950
     
PCU1
 
0.833
     
Competitor uncertainty (CU)

CU2
  
0.952
    
CU3
  
0.944
    
CU1
  
0.668
    
Government policy uncertainty (GU)

GU1
   
0.825
   
GU3
   
0.800
   
GU2
   
0.758
   
Process uncertainty (PU)

PU2
    
0.824
  
PU1
    
0.771
  
PU3
    
0.714
  
Supply uncertainty (SU)

SU1
     
0.840
 
SU2
     
0.808
 
SU3
     
0.678
 
Demand uncertainty (DU)

DU2
      
0.832

DU3
      
0.727

DU1
      
0.700

Eigen value

3.270

2.706

2.393

2.085

2.004

1.989

1.978

% of variance

14.864

12.298

10.879

9.478

9.108

9.039

8.989

Cumulative % of variance

14.864

27.161

38.041

47.519

56.627

65.667

74.656

 Extraction method: principal component analysis
 Before we conduct a principal component analysis or factor analysis, we must verify if the necessary conditions are fulfilled:
 To measure the scale reliability, we calculate the correlation matrix of the 22 uncertainty factors and the determinant. Since the determinant is different from zero, the factor analysis may be completed. Moreover, in order to measure scale reliability of the questionnaire, Cronbach’s alpha is used (Bryman 2003; Haire et al. 1995). The value of Cronbach’s alpha is accepted for an exploratory study if it exceeds 0.7 (Nunnally 1967). The Cronbach’s alpha of these scales ranges from 0.710 to 0.922. No items are deleted in the analysis
 The scale validity is measured by the KaiserMeyerOlkin Measure (KMO) and Bartlett’s Test. The result for the KaiserMeyerOlkin Measure (KMO) is acceptable since it is larger than 0.6 (Kaiser 1974), and Bartlett’s Test is highly significant at p < 0.000. Scale validity indicates the construct is able to measure accurately the concept under study (Haire et al. 1995)
 The construct validity is measured by explanatory factory analysis (EFA) (Haire et al. 1995). All components have Eigenvalues larger than 1, which confirms the construct validity
 Rotation method: varimax with Kaiser normalization
 Source: own data (2017) and SPSS