MULTIDIMENSIONAL POVERTY MAPPING

Abstract

Previously in Pakistan, the research on poverty has been done on the basis of income (money metric poverty) e.g. if a person is earning below the minimum income level then they are categorized as being poor. Multidimensional poverty mapping is based on various deprivations such as a lack of access to health facilities, lack of access to recreational spots, lack of access to markets, the inability to attain a good education and a poor standard of living which involves poor quality of housing and road infrastructure (Li & Weng, 2007). In Pakistan, the poverty mapping on the basis of indicators has been done on a huge level e.g. provincial and district wise poverty mapping. In our project, the main focus is to identify poor regions on a small level e.g. city level using high-resolution satellite images along with some proxy data of health, education and an average number of people per household. Different indicators like access to health facilities, access to educational institutions, access to markets, access to parks, student attendance in schools and the average number of people per household are used (F AO., 2002) along with the classification results of high resolution satellite images which display the good quality housing areas, medium quality housing areas and the bad quality housing areas to predict the areas which are more prone to poverty. Thus, an area is poor if the housing quality is poor; there are unordered and unequal houses, an average number of people in a household is too much and if it has poor access to health, education facilities, markets, and parks. The salient feature of our project is that we won't be taking income into account and the keynote is that if a person does not have access to the basic necessities of life e.g. health, education, commercial areas, recreational areas and if they live in a densely populated area with unordered and unequal houses and a lot of people living in one house than their living standard is not good and they are poor.