Page 140 - Hojnik, Jana. 2017. In Persuit of Eco-innovation. Drivers and Consequences of Eco-innovation at Firm Level. Koper: University of Primorska Press
P. 140
In Pursuit of Eco-innovation
SEM has also the ability to incorporate latent variables into the anal-
ysis. A latent variable, or latent construct, is a hypothesized and unob-
served concept that can be represented by observable or measurable vari-
ables. It is measured indirectly by examining consistency among multiple
measured variables, sometimes referred to as manifest variables, or indi-
cators, which are gathered through various data collection methods (e.g.,
surveys, tests, observational methods) (Hair et al. 2009). The standard
method of estimating free parameters in SEM is to employ maximum
likelihood (ML). A growing body of research indicates that ML per-
forms reasonably well under a variety of less-than-optimal analytic condi-
tions (e.g., small sample size, excessive kurtosis) (Hoyle and Panter 1995).
Moreover, Hair et al. (2006) pointed out that several readily available sta-
tistical programs are convenient for performing SEM. Traditionally, the
140 most widely used program is LISREL. EQS is another widely available
program that also can perform regression and factor analysis and can test
structural models. AMOS is a third program that has gained popularity
because it is user-friendly and available as an addition to SPSS.
Evaluation of the results
For all the constructs measured in this survey, we first conducted an ex-
ploratory factor analysis, which is a class of procedures primarily used for
data reduction and summarization (Malhotra 1993). Our main aim was
to see how many factors are extracted based on the variables that were
used to measure different constructs.
When evaluating exploratory factor analysis, the key statistics associ-
ated with factor analysis are as follows (Malhotra 1993):
- Bartlett’s test of sphericity is a test statistic used to examine the
hypothesis that the variables are uncorrelated in the population.
In other words, the population correlation matrix is an identity
matrix in which each variable correlates perfectly with itself (r =
1) but has no correlation with the other variables (r = 0).
- Correlation matrix is a lower triangle matrix showing the simple
correlations between all possible pairs of variables included in
the analysis.
- Communality is the amount of variance a variable shares with all
the other variables being considered. This is also the proportion
of variance explained by the common factors.
- Eigenvalue represents the total variance explained by each fac-
tor.
SEM has also the ability to incorporate latent variables into the anal-
ysis. A latent variable, or latent construct, is a hypothesized and unob-
served concept that can be represented by observable or measurable vari-
ables. It is measured indirectly by examining consistency among multiple
measured variables, sometimes referred to as manifest variables, or indi-
cators, which are gathered through various data collection methods (e.g.,
surveys, tests, observational methods) (Hair et al. 2009). The standard
method of estimating free parameters in SEM is to employ maximum
likelihood (ML). A growing body of research indicates that ML per-
forms reasonably well under a variety of less-than-optimal analytic condi-
tions (e.g., small sample size, excessive kurtosis) (Hoyle and Panter 1995).
Moreover, Hair et al. (2006) pointed out that several readily available sta-
tistical programs are convenient for performing SEM. Traditionally, the
140 most widely used program is LISREL. EQS is another widely available
program that also can perform regression and factor analysis and can test
structural models. AMOS is a third program that has gained popularity
because it is user-friendly and available as an addition to SPSS.
Evaluation of the results
For all the constructs measured in this survey, we first conducted an ex-
ploratory factor analysis, which is a class of procedures primarily used for
data reduction and summarization (Malhotra 1993). Our main aim was
to see how many factors are extracted based on the variables that were
used to measure different constructs.
When evaluating exploratory factor analysis, the key statistics associ-
ated with factor analysis are as follows (Malhotra 1993):
- Bartlett’s test of sphericity is a test statistic used to examine the
hypothesis that the variables are uncorrelated in the population.
In other words, the population correlation matrix is an identity
matrix in which each variable correlates perfectly with itself (r =
1) but has no correlation with the other variables (r = 0).
- Correlation matrix is a lower triangle matrix showing the simple
correlations between all possible pairs of variables included in
the analysis.
- Communality is the amount of variance a variable shares with all
the other variables being considered. This is also the proportion
of variance explained by the common factors.
- Eigenvalue represents the total variance explained by each fac-
tor.
