Using logistic regression to determine the major causes of the rural poverty a case study of Kenya
Abstract
Poverty is a multi-dimensional phenomenon whose causes are complex and
difficult to Measure precisely. The need to articulate causes of poverty in
developing countries is Paramount given its effects. This study examines the
major causes of rural poverty from 942 households. For poverty reduction
strategies to be effective, they must be closely linked with the objectives of
the poverty alleviation program. Most of the official poverty statistics released
including the data statistics used in this study focuses on the poverty
incidence, the income gap, and the poverty gap and traps. However, identifying
the characteristics of vulnerable families and the vulnerable provinces
is important towards understanding the causes of their vulnerability and in
formulating programs on poverty reduction. This study proposed a framework
and a methodology for identifying the sub locations, districts and the
provinces where these families are located and the indicators of a potential
deprived community to guide program implementers in the allocation
of resources. In this study logistic regression analysis was applied. The results
indicated that family size, geographical location, agriculture knowledge,
empowerment and the gender of the head of the household as the key determinant
of the probability of one being poor in a rural set up. Age is factor
when run as a univariate logistic regression but when run in a multivariate
logistic regrassion it is not a predictor of ones probability of being poor.
Citation
M.Sc (Social Statistics)Sponsorhip
University of NairobiPublisher
School of Mathematics, University of Nairobi
Description
Master of Science Thesis