I am trying to calculate a difference-in-difference estimate. I have a binary outcome variable, and have estimated risk differences for my intervention and control groups.
If we take an example where:
- The intervention group has a risk difference of 0.03724 (i.e. 3.7% increase in the risk of the outcome) with a standard error of 0.01360
- The control group has a risk difference of 0.00410 with a standard error of 0.00316
Then we can calculate the difference-in-difference is 0.03724 - 0.00410 = +0.03314. What is the standard error of this estimate?