Spatial analysis of poverty among women in Ghana
Loading...
Date
October 15, 2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In Ghana, although women constitute more than half of the total population, they
are grossly disadvantaged in the sharing of political power, wealth, in
uence, employments
etc. In some parts of the country, the well-being of women is highly
affected by these inequities. This study used spatial econometric techniques to
investigate the spatial distribution of poverty among women and explored the
main factors that determine the incidence of women poverty in Ghana.This study
draws on the 2008 GDHS household's women data with their wealth factor scores
(economic status) as the response variable. PCA technique was employed to
obtain 13 components from an initial 27 socio-economic, demographic and geographic
variables. These were the regression variables in GeoDa. The results
show a highly positive significant Moran's I (I= 0.396; p=0.001), indicating that
neighboring areas have similar poverty status; that is poverty is a geographical
phenomenon. Also, the poverty maps identifed the three northern regions as the
most endemic women poverty areas. The regression analysis further indicated
that SLM [R2 = 79:2%;Log.L=-234;AIC=498;SC=558] fit the data better than
the SEM [R2 = 79:0%;Log.L=-242;AIC=511;SC=568] and OLS [R2 = 74:0%
;Log.L=-270;AIC=569;SC=630] models. The selection of SLM indicates that the
rate of poverty of one area affects the poverty rates of its neighbors. The major
significant determinants of women poverty in 2008 were the education related
variables, parity, some occupational related variables and marital status (married).
The variables that had no significant relationships with poverty are female
headed household and couple married without living together.
Description
A thesis submitted to the Department of Mathematics,
Kwame Nkrumah University of Science and Technology in
partial fulfillment of the requirement for the Degree
of Master of Philosophy in Mathematical Statistics.