jocain/Liberia-Education-Prediction

Classification Analysis of Liberian Demographic Data ### Description In this project, I explore a variety of classification methods as a means of modeling education levels in a set of data concerning Liberian demographic information. ### Data Overview The data used in this project was collected in the Liberia Demographic and Health Survey of 2013. The survey was conducted by the Liberia Institute of Statistics and Geo-Information (LISGI) from mid-March to mid-July in 2013 and was funded by a number of international government and non-government organizations, including USAID and UNICEF. The survey covered a broad stroke of information ranging from HIV testing to access to drinking water. For the purposes of this project, Location, Size, Wealth, Gender, and Age were used as features in an attempt to model a target variable, Education. Location and Gender are nominal variables, Age and Size are treated as continuous variables, and Wealth and Education ordinal variable. Location refers to six groups of counties partitioned by LISGI, designated as the North Western, North Central, South Central, South Eastern A, and South-Eastern B regions, shown in the map below: The survey was conducted such that the random houses in each of the regions were chosen, then all the members of the household answered the survey. Therefore, while Gender, Age, Education refer to an individual's qualities, Location, Size, and Wealth refer to attributes of the household they belong to. Household wealth in this study is broken up into quintiles, and Size refers to the number of members of a household. Of those members, Gender is taken to be binary (male or female), Age is the member's age, and Education is ranked by the highest level of education the member has received: None (0), Preschool/Kindergarten (1), Primary (2), Secondary (3), and Tertiary (4). This particular variable required some cleaning and amending. The original mapping of the survey responses was the following: None was 0, Preschool/Kindergarten was 6, Primary was 1, Second …