Page 780 - ISC PROCEEDINGS 21.4

Table 4. Rotated Component Matrix a
Component
1 2 3 4
DI1 .709
DI2 .814
DI3 .846
DI4 .713
DL1 .723
DL2 .635
DL3 .769
DL4 .770
DC1 .817
DC2 .800
DC3 .856
DC4 .859
DA1 .692
DA2 .785
DA3 .843
DA4 .840
Source: Author
Varimax rotation with Kaiser normalization converged after six iterations, indicating
a stable factor structure. The rotated component matrix shows that all observed variables
load onto four distinct factors, consistent with the proposed theoretical model. All factor
loadings exceed 0.6 (Hair et al.), confirming good convergent validity. Specifically, DC
items load strongly on Factor 1 (0.800–0.859), DA items on Factor 2 (0.692–0.843), DI
items on Factor 3 (0.709–0.846), and DL items on Factor 4 (0.635–0.770), reflecting the
four dimensions of digital transformation. No significant cross-loadings are observed,
indicating clear factor separation and strong discriminant validity. The results confirm that
the measurement scales are consistent with the theoretical framework and meet the
requirements for subsequent regression analysis.
Factor analysis for the dependent variable
In addition to analyzing the independent variables, the study also conducted
exploratory factor analysis for the Sustainable Human Resource Management (SHRM)
scale to evaluate the convergence of the observed variables representing this construct.
Table 5. Rotated Component Matrix for the Dependent Factor
Observable Variable Factor Loading
SHRM1 .617
SHRM2 .680
SHRM3 .864
Source: Author
Exploratory factor analysis was conducted for the Sustainable Human Resource
Management (SHRM) scale. The results show that all three observed variables (SHRM1,
SHRM2, SHRM3) load on a single factor, with loadings of 0.617, 0.680, and 0.864,
respectively, all exceeding the threshold of 0.5. These findings confirm good convergent
validity, with SHRM3 contributing most strongly to the construct. The results also indicate
that the scale effectively captures key aspects of SHRM, including long-term development,

779