AD ALTA
JOURNAL OF INTERDISCIPLINARY RESEARCH
For the purpose of factor analysis the value of Kaiser-Meier-
Olkin test should reach the value of at least 0.5 (value range is
between 0 and 1). In order to assess whether it is possible to use
the factor analysis, Kaiser-Meyer-Olkin method (KMO) and
Bartlett’s test of sphericity have been used. The KMO method is
based on selective correlation and partial correlation coefficients.
For the indicators in factor analysis KMO are observed
according to the highest level of acceptance, which means that
the performed level of usefulness of the factor analysis reaches
high value. Bartlett’s test of sphericity is a statistic test used to
examine the hypothesis that the variables are correlated or
uncorrelated.
Value for KMO test was reached by 0,764 and for Bartlett’s test
by 0,000. Therefore, factor analysis itself could be applied. The
total variance of the performance indicators is explained by
means of eigenvalues, representing the total variance explained
by each factor. The eigenvalues show that only three items have
reached the minimum value of 1. From this point of view,
Extraction Sums of Squared Loadings with cumulative
percentage are important. Factor analysis has extracted different
numbers of factors, which explains variances of all cases
(81,54%).
Table 3 Results of factor analysis
Cronbach’s
alpha
Factor
1
Cash flow
0,994
0,779
ACCEPT
P/L for period (Net
income)
0,994
Operating revenue
0,987
Shareholder funds
0,973
Factor
2
ROA using P/L before
tax
0,813
0,437
NOT
ACCEPT
ROE before tax
0,857
Source: own work by authors
Results of factor analysis provide in two factors, from which are
acceptable value of Cronbach’s alpha only for one of them. Last
factor has Cronbach’s alpha value under minimal acceptable
value (under 0,500). Final values calculating acceptable factor
need the transformation of individual coefficients. These
coefficients express significance of the used elements. Their sum
total must be 1. The individual factor indices have been defined
by the procedures as follow:
Value of this factor can be calculated for the individual outdoor
producer and on the basis of their results a list of businesses can
be compiled. Indices can determine important factors of
business, playing the key role in achieving the set of objectives.
Proposed financial performance indicators should help
companies to demonstrate a progress towards the objectives of
sustainability. Also we can see basic statistics of observed
indexes in Table 4.
Table 4 Descriptive statistics of observed factor
Mean
Median
Variance
Std.
deviation
Factor 1
2016,6704
165,8953
295000,99809
870,08888,3
Factor 1
-
grouped
3
3
1,414
2
Source: own work by authors
Pivot tables have been employed to find possible dependencies
between observed factors and company size and region of
company, for results of the dependency tests see Table 5. Results
of the dependence examination in individual variable categories
are depicted in the following results of Pearson’s chi-square test.
Maintaining the % reliability of the test, the values for
connection between individual factors and company size have
been determined within 0.05, which represents 5% reliability
level. Established values of Pearson’s test for the variables are
showed in Table 5 (i.e., less than 0.05). Therefore, that bring us
to the conclusion that an alternative hypothesis is applied – there
are dependencies between all observed factors and company size
for all observed indexes. Past results have revealed the
relationship between indexes and company size and European
region. Subsequently, degree of such dependence has been
examined. To that end, the intensity of dependence determined
by means of contingency coefficient.
Table 5 Pearson’s test of the relationship between individual
indexes, company size and European region for observed factor
Corporate size
European region
Value
Signif.
Value
Signif.
Pearson χ
2
2010,429
0,000
225,020
0,000
Contingency
coefficient
0,555
0,208
Source: own work by authors
The intensity of dependence ranges between
〈0;1〉. That means
that the higher the absolute value, the greater the intensity of
dependence. Table 5 shows that observed factor is close
connected within the size of the company and region of Europe –
all significance values are in 5% of limit of error. Intensity of the
dependence is given by Contingency coefficient, which provides
view in this connection. All four defined connection between
observed indexes and corporate size and region reach accurate
values and there are confirmed dependency between them.
Load indicators (Mass) indicate load line which represents the
percentage of information across the table in appropriate
category. That loads are obtained as the ratios of the row and
column marginal frequencies (n
i+
, n
+j
) in whole table of
individual categories (n).
Score in dimension describes individual variables score in two
main dimensions. These dimensions don‘t represent any specific
area, because they are reduced to from multi-dimension space.
All data in rows and columns have been usually in multi-
dimension space, which are reduced into two. Providing
information of raw data has not been modified after multi-
dimension space reduction of these variables. Inertia indicator
represents the share comprehensive information on the profile
(on the relevant point). This characteristic is independent of the
number of dimensions. Corresponding map includes a graphical
representation of both row and column categories according to
their dimension scores (Hebák et al., 2007; D’Esposito, de
Stefano, Ragozini, 2014).
- 146 -