6 Approach to Vulnerability Assessment
Vulnerability is often reflected in the state of the economic system as well as the
socio-economic features of the population living in that system. The current section
of the report attempts to build a picture of the socio-economic context of vulnerability by focusing on indicators that measure both the state of development of the
people as well as its capacity to progress further. In addition, attempt has been made
to construct a vulnerability index for each district of Karnataka and rank them in
terms of their performance on the index. The index attempts to capture the comprehensive scale of vulnerability by considering some of the key indicators (that
serve as proxies) for the assessment (Table 3).
There is consensus among researchers to address vulnerability issues at the
regional level (Hiremath and Shiyani 2013). Therefore districts have been taken as a
unit for developing vulnerability indices. In the next step we have selected
important indicators for the vulnerability assessment. After the selection of indicators, data pertinent to the selected indicators were compiled (Table 3). In the next
step, a Principal component analysis (PCA) was conducted to identify variability
among the selected variables and finally vulnerability indices were developed.
Figure 14 presents the details of the method adopted for assessment of vulnerability
across the districts of Karnataka.
6.1 Principal Component Analysis (PCA)
A PCA was conducted to identify the variability among selected variables (indicators) for this study. PCA is a data reduction methodology that identifies smaller
number of components that explains most of the variance observed in the larger
data set. The goal is to arrive at a minimum number of components that will
adequately account for the covariation among the larger number of analysis variables. PCA is a tool that converts a number of potentially correlated variables into a
set of uncorrelated variables that capture the variability in the underlying data. It is a
statistical method that thus transforms a given data set to a smaller number of
uncorrelated variables called principal components (PCs). The first PC accounts for
a large share of variability in the data, and each succeeding component accounts for
as much of the remaining variability as possible. PCA approach provides several
potential advantages in the aggregation of spatially explicit, potentially incommensurable variables. When the original variables are correlated then the higher
order PCs will capture more of the total variability in the data than any individual
original variable. Excluding the lower order PCs reduces the dimensionality
(number of variables) of the data while minimizing the loss of information (Smith
2002). PCA thus helps reduce from a large number of individual indicators to a
small number of composite, unitless indices (PCs) while reducing the trade-off
between richness of information and communicability.
178
K.V. Raju et al.
Vulnerability is often reflected in the state of the economic system as well as the
socio-economic features of the population living in that system. The current section
of the report attempts to build a picture of the socio-economic context of vulnerability by focusing on indicators that measure both the state of development of the
people as well as its capacity to progress further. In addition, attempt has been made
to construct a vulnerability index for each district of Karnataka and rank them in
terms of their performance on the index. The index attempts to capture the comprehensive scale of vulnerability by considering some of the key indicators (that
serve as proxies) for the assessment (Table 3).
There is consensus among researchers to address vulnerability issues at the
regional level (Hiremath and Shiyani 2013). Therefore districts have been taken as a
unit for developing vulnerability indices. In the next step we have selected
important indicators for the vulnerability assessment. After the selection of indicators, data pertinent to the selected indicators were compiled (Table 3). In the next
step, a Principal component analysis (PCA) was conducted to identify variability
among the selected variables and finally vulnerability indices were developed.
Figure 14 presents the details of the method adopted for assessment of vulnerability
across the districts of Karnataka.
6.1 Principal Component Analysis (PCA)
A PCA was conducted to identify the variability among selected variables (indicators) for this study. PCA is a data reduction methodology that identifies smaller
number of components that explains most of the variance observed in the larger
data set. The goal is to arrive at a minimum number of components that will
adequately account for the covariation among the larger number of analysis variables. PCA is a tool that converts a number of potentially correlated variables into a
set of uncorrelated variables that capture the variability in the underlying data. It is a
statistical method that thus transforms a given data set to a smaller number of
uncorrelated variables called principal components (PCs). The first PC accounts for
a large share of variability in the data, and each succeeding component accounts for
as much of the remaining variability as possible. PCA approach provides several
potential advantages in the aggregation of spatially explicit, potentially incommensurable variables. When the original variables are correlated then the higher
order PCs will capture more of the total variability in the data than any individual
original variable. Excluding the lower order PCs reduces the dimensionality
(number of variables) of the data while minimizing the loss of information (Smith
2002). PCA thus helps reduce from a large number of individual indicators to a
small number of composite, unitless indices (PCs) while reducing the trade-off
between richness of information and communicability.
178
K.V. Raju et al.