Empowerment and Domestic Violence

contrast to other blocks in the district of Bankura. Second, two Grampanchayats of

this block, namely Mirjapur and Madanmohanpur have been selected randomly. In

the third step we have selected three villages from each selected Grampanchayat.

Finally, from the selected villages we have randomly selected the sample

households and interviewed on the basis of a structured questionnaire during

2010. Our sample has covered the information of 125 ever-married women from

125 households.

In order to study the nature of economic empowerment of the rural women we

have considered nine indicators of economic empowerment listed below.

¾ The woman has own asset.

¾ The woman is earner.

¾ The woman has bank account.

¾ The woman uses her own asset.

¾ The present occupation of the woman is secure.

¾ The woman contributes in family expenditure.

¾ The woman spends independently for family.

¾ The woman takes expenditure/ investment/saving decision.

¾ The woman has access to formal credit.

Wehavescaledtheseitemsby‘1’ ifthewomanrespondspositivelyfortheparticular

decisive factor and by ‘0’ if she does not. We have constructed the Composite

Index of Economic Empowerment for the sample women using the technique of

Principal Component Analysis (PCA) on the selected indicators. This technique

develops a set of uncorrelated variables called Principal Components (PCs) from

a set of indicators or variables. This study has extracted Principal Components by

Kaiser Criterion which considers only the components having Eigen value greater

than one. However, in case of PCA without rotation, the eigenvectors may not

align close to the data clusters and thus may not focus the actual states as well.

The rotated PCA methods rotate the PCA eigenvectors so that they align closer

to the cluster of data. Applying Varimax rotation strategy, which maximizes the

variance of the rotated squared PCs, we have rotated the PCA specifying the fixed

number of components. Respective rotated component scores have been obtained

by regression method. Finally, the Composite Index of Economic Empowerment

for the sample women has been calculated taking the weighted sum of component

scores. The weight of a particular PC is the percentage of variations in the data set

explained by the respective PC after rotation.As the incidence of domestic violence

is a dichotomous variable we have formulated a probit model for assessing the

435