Hello Assalamualaikum, I am Fawad Ahmad, and I will be explaining the methodology
1. Blinder-Oaxaca Decomposition
we used this technique called Blinder-Oaxaca Decomposition. It is widely used to study
the labor market outcomes by different groups such as gender, race, region etc.
this technique helps us calculate the gap which is further divided into 2 parts. One is the
explained part and the other one is the unexplained part.
In our case It is the gap between wages of males and females.
The explained part gives us the difference in wages due to the difference in
productivities caused by the selected variables.
Whereas the unexplained part gives us the difference in wages due to gender
discrimination + the variables we haven’t included in the model yet
We can increase the explained part by adding more variables into the model
2. Model
Moving forward we have our model
This is the standard regression equation for male and female income
α -> intercept
X -> variable
Β-> coefficient
u-> residual part
we subtract the log income of males from log income of females, this gives us Gender
wage gap
this is done through the technique mentioned earlier
we use log to achieve a more normally distributed data with constant variance rather than
skewed data.
3. Frequency Tables
Here are some frequency tables of the variables that we are using in our model. They also
contain the total number of observations.
In gender 0 is coded as males, 1 is for females
Same here in table 2 for vocational training
In marital status 0 is for unmarried and 1 is for married labor force workers
For region, 0 is coded as urban and 1= rural
And for migration, 0 is rural to urban and 1 is urban to rural
Next up we have made this variable of log wage which was created after taking
the log of the variable total wage
Moving on we have our findings
4. Findings
Overall Decomposition
Group 1 is male workers and Group 2 is female workers. Their mean log wages
are 12.41 and 12.09 respectively
If we subtract the wages of group 1 and group 2 we get the gender wage gap
which is 0.3225
Out of which 44.1% is the explained part and 55.9%
Determinants
Age is positively impacting the wage gap
Experience is negatively impacting the wage gap
Marital Status is positively impacting the wage gap
Education Level is negatively impacting the wage gap
Vocational Training is positively impacting the wage gap
Household head is positively impacting the wage gap
Region is positively impacting the wage gap
Employment Type is negatively impacting the wage gap
Hours of work in a week is negatively impacting the wage gap
Migration is negatively impacting the wage gap
Now Nimra will come again and explain the conclusion and policy recommendations
5. Defining Variables
socio economic and demographic variables
o age
o marital status
o education level
o technical training
o Region
