Hi everybody!
I am running a diff-in-diff analysis in Stata:
Female employment in the service sector = treatment dummy + period dummy + interaction of both
--> Female employment in services is measured as a share of total female employment --> so from 0 to 100%
--> the period dummy equals one for all years after 2015
If my interaction coefficient is, let's say, 1.23 and significant, is the following interpretation correct?:
"Relative to the control group, female employment in the service sector in the treatment group experiences an additional increase of 1.23 percentage points after 2015."
Now, I want to add a triple interaction --> treatment dummy * period dummy * natural log of private sector investment
What would the interpretation now be? The dependent variable is expressed in percentage and the triple interaction consists of two dummies and the natural logarithm of investment, which otherwise is expressed in millions of UDS.
Can anyone help?
Thanks!
I am running a diff-in-diff analysis in Stata:
Female employment in the service sector = treatment dummy + period dummy + interaction of both
--> Female employment in services is measured as a share of total female employment --> so from 0 to 100%
--> the period dummy equals one for all years after 2015
If my interaction coefficient is, let's say, 1.23 and significant, is the following interpretation correct?:
"Relative to the control group, female employment in the service sector in the treatment group experiences an additional increase of 1.23 percentage points after 2015."
Now, I want to add a triple interaction --> treatment dummy * period dummy * natural log of private sector investment
What would the interpretation now be? The dependent variable is expressed in percentage and the triple interaction consists of two dummies and the natural logarithm of investment, which otherwise is expressed in millions of UDS.
Can anyone help?
Thanks!
