Using the Logistic Regression Model to analyze the Determinants of Poverty at the Household Level in Erbil Governorate for the Years (2012 And 2018)

Abstract

In order to diagnose the factors affecting poverty in Erbil Governorate for the years (2012 and 2018), poverty regression was conducted as a binary dependent variable, which takes two values: (Y = 1) if the family is poor and (Y = 0) if it is not. It lacks (9) explanatory variables affecting it. They are (family size, number of working members in the family, age of the head of the family, educational level of the head of the family, dependency ratio, housing ownership, geographical location, gender of the head of the family, , which takes a value of (0) if the observation dates back to 2012, and a value (1) for the year 2018) for the study sample using a logistic regression model After that, the chosen model was subjected to theoretical, statistical and measurement criteria to analyze and evaluate the results obtained. It has been shown that increasing the size of the family, the age of the head of the family, the dependency ratio, families renting a house, families living in the countryside, and families headed by women when the observation returns to 2018 have a negative impact on the probability of families falling into poverty.  That is, these determinants contribute to an increase in the incidence of poverty, while the results showed that each of the determinants: the number of working members in the family and the educational level of the head of the family has a positive effect on reducing the incidence of poverty, meaning that they contribute to reducing poverty for the study sample.

The values ​​of both () and () indicate that approximately (0.28) and (0.77) respectively, of the changes in the dependent variable are due to changes in the explanatory variables, and the rest of the changes in the dependent variable are due to a group of factors not included in this estimated model.