Page 230 - Hojnik, Jana. 2017. In Persuit of Eco-innovation. Drivers and Consequences of Eco-innovation at Firm Level. Koper: University of Primorska Press
P. 230
In Pursuit of Eco-innovation
measured by the extent of sales on foreign markets, ranging from 0 to 100% (level 1 = 0%; le-
vel 2 = 1-10%; level 3 = 11-20%; level 4 = 21-30%; level 5 = 31-50%; level 6 = 51-70%; level 7 = 71-
90%; level 8 = 91-100%); variable Number of operation modes was constructed by summing
all operation modes); St. Dev. = standard deviation; Skew = skewness; St. Err. of Skew = stan-
dard error of skewness; Kurt = kurtosis; St. Err. Kurt = standard error of kurtosis.
Further, we conducted an exploratory factor analysis (the method
of extraction was the Maximum Likelihood Method, while the select-
ed rotation was Direct Oblimin rotation). Before the analysis, all meas-
urement items were checked for normality of distribution (see Table 79).
Results have shown that the ratio of standard errors of kurtosis and skew-
ness range between values of -2 and 2, which implies normality of distri-
bution.
The appropriateness of factor analysis was determined by examining
230 the correlation matrix of economic benefit items. The Bartlett’s test of
sphericity showed that the correlation matrix has significant correlations
(p < 0.05). Furthermore, the Kaiser-Meyer-Olkin measure of sampling
adequacy was 0.514, which indicates a sufficient sample adequacy.
After consideration of each item’s communality index and its contri-
bution, we retained all the items (the lowest communality after extrac-
tion was 0.124). In the process of analysis, researchers usually delete or
exclude the items that have low communalities after extraction (below
the threshold of 0.20). However, we retained the item “Total number of
operation modes” in further analyses despite the low communality after
extraction, because of its relevance. The number of expected factors was
one, and the extracted factor was one, explaining 50.377% of variance.
Table 80: KMO and Bartlett’s test of sphericity (Internationalization)
KMO and Bartlett’s test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.514
93.843
Approx. chi-square
3
Bartlett’s test of sphericity df 0.000
Sig.
A confirmatory factor analysis was conducted in order to validate the
findings of the exploratory factor analysis, which resulted in one factor
composed of three items. This has also been confirmed by the confirm-
atory factor analysis. The construct internationalization comprises three
items. All the coefficients were found to be positive, high and significant,
and are indicated in Table 81 and Figure 25.
measured by the extent of sales on foreign markets, ranging from 0 to 100% (level 1 = 0%; le-
vel 2 = 1-10%; level 3 = 11-20%; level 4 = 21-30%; level 5 = 31-50%; level 6 = 51-70%; level 7 = 71-
90%; level 8 = 91-100%); variable Number of operation modes was constructed by summing
all operation modes); St. Dev. = standard deviation; Skew = skewness; St. Err. of Skew = stan-
dard error of skewness; Kurt = kurtosis; St. Err. Kurt = standard error of kurtosis.
Further, we conducted an exploratory factor analysis (the method
of extraction was the Maximum Likelihood Method, while the select-
ed rotation was Direct Oblimin rotation). Before the analysis, all meas-
urement items were checked for normality of distribution (see Table 79).
Results have shown that the ratio of standard errors of kurtosis and skew-
ness range between values of -2 and 2, which implies normality of distri-
bution.
The appropriateness of factor analysis was determined by examining
230 the correlation matrix of economic benefit items. The Bartlett’s test of
sphericity showed that the correlation matrix has significant correlations
(p < 0.05). Furthermore, the Kaiser-Meyer-Olkin measure of sampling
adequacy was 0.514, which indicates a sufficient sample adequacy.
After consideration of each item’s communality index and its contri-
bution, we retained all the items (the lowest communality after extrac-
tion was 0.124). In the process of analysis, researchers usually delete or
exclude the items that have low communalities after extraction (below
the threshold of 0.20). However, we retained the item “Total number of
operation modes” in further analyses despite the low communality after
extraction, because of its relevance. The number of expected factors was
one, and the extracted factor was one, explaining 50.377% of variance.
Table 80: KMO and Bartlett’s test of sphericity (Internationalization)
KMO and Bartlett’s test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.514
93.843
Approx. chi-square
3
Bartlett’s test of sphericity df 0.000
Sig.
A confirmatory factor analysis was conducted in order to validate the
findings of the exploratory factor analysis, which resulted in one factor
composed of three items. This has also been confirmed by the confirm-
atory factor analysis. The construct internationalization comprises three
items. All the coefficients were found to be positive, high and significant,
and are indicated in Table 81 and Figure 25.
