Page 207 - Contributed Paper Session (CPS) - Volume 3

CPS1996 Ali Hebishi Kamel A. et al.
Inequality may refer to either difference within a group (such as income
inequality). Disparity specifically refers to differences between groups (or
between a group and a reference value). In this study we will focus difference
on income and completed education, so it is the answer for the third research
question. Also how the disparities in level of education by completed and
uncompleted go to disparities in their incomes.

4. Result
• We used log-transform for income instead of total income to express more
variation in models.
• All simple regression models for independent variable included in the
income model is statistically significant separately with household income
(log income) the same for education model (binary logistic) which reflects
the nature of Egyptian population.
• All independent variables in Multipule Regression model of income
reflects statistically significant relationship between log(income) and each
of education level of HH, household size, place of residence, disability
status of HH, sex of HH and Age of HH with different values of level of
significant (1% , 5% , 10%) and have positive relationship between income
and all independent variables and interactions expect HH disablity status.
• For household income model the best fit model for simple regression
when independent variable is education level of HH because it has the
2
lowest AIC (the tool of control the best fit model) and High adjust R ; Also
the best model for Education that has the relation between level of
education individual (completed secondary and under secondary) and
household income with the lowest AIC and that reflect strong relationship
between income and education.
• Higher Level of education of HH increases household income (dependent
variable), because if HH completed secondary education that have income
higher than under secondary education as (independent variable) alone in
simple regression model (Figure(1)). And also when we add some
independent variables for the model the level of education of HH have
high effect on income from the results of this model (the least AIC in the
model), So it answer the first research question "Is THE RELATION
BETWEEN INCOME OF HOUSEHOLD AND EDUCATION?”.
• In Education model (Binary logistic regression): there is a positive
relationship between Completed secondary education and income and all
other indepandant variables such as gender, place of residence and
mother education except HH poverty status and interaction included that
variable.
• Household Income influence on completed education of individual,
because when we have increase in income there is an increase in the odd

196 | I S I W S C 2 0 1 9