Page 199 - Hojnik, Jana. 2017. In Persuit of Eco-innovation. Drivers and Consequences of Eco-innovation at Firm Level. Koper: University of Primorska Press
P. 199
Results

Table 56: Eco-innovation dimension’s scale convergence – summary for all three eco-innova-
tion dimensions and eco-innovation construct

Dimension Cronbach’s Range of NFI Model fit indexes
N. of items alpha reli- standard- NNFI CFI SRMR RMSEA
ized coeffi-
ability
cients*

Product eco-in- 4 0.872 0.60 to 0.80 0.993 0.992 0.997 0.017 0.05
novation

Process eco-in- 5 0.912 0.72 to 0.92 0.964 0.939 0.970 0.036 0.15
novation

Organizational 6 0.956 0.80 to 0.93 0.945 0.918 0.951 0.030 0.19
eco-innovation

Eco-innovation 15 0.952 0.62 to 0.93 0.928 0.945 0.954 0.044 0.08
construct
199

Note: N. of items = number of items of each eco-innovation dimension; * all standardized
coefficients are positive, high and significant (sig. < 0.05).

Convergent and discriminant validity of the eco-innovation construct
The eco-innovation dimensions were tested for convergent and discrimi-
nant validity together in the eco-innovation construct structural model,
where dimensions were modeled as first-order latent constructs and cor-
related with each other (see Figure 16). The model showed good fit (NFI
= 0.928; NNFI = 0.945; CFI = 0.954 – over the threshold of 0.90; SRMR
= 0.044; RMSEA = 0.086 – below the threshold of 0.10). Moreover, all
coefficients were found to be positive, high and significant. Also, the reli-
ability coefficients (Cronbach’s alpha = 0.952; RHO = 0.968, were high –
above the threshold of 0.70).

Figure 16 illustrates the standardized solution for the eco-innovation
construct, composed of the three dimensions of product, process and
organizational eco-innovation. The 15 items of these three dimensions
measure the entire eco-innovation construct. Table 57 offers results per-
taining to the standardized coefficients and their squares for all items of
the eco-innovation construct.
   194   195   196   197   198   199   200   201   202   203   204