Dear All,

I am trying to estimate the causal effect of maternal education on a range of outcomes using fuzzy RDD. I am using the exposure to the compulsory education law as an instrument for individuals's schooling.

However, I could not find the appropriate syntax for me.

Here is my setting:

Compulsory education law (increased madatory education from 5 to 8 years) was binding for those who are born in and after 1987. However, individuals born before 1987 were not affected. Yet, the situation is fuzzy for those who born in 1986. This allows me to implement fuzzy RDD.

My dependent variable---> child health
My treatment variable ---> To understand whether they are treated by the educational reform, I use individuals' year of birth. Then, I created reform_exposure variable ---> 1=if born in 1987 and further; 0= if born before 1987
My education variable ---> 1= if an individual completed at least 8 years of education, 0= otherwise

For Fuzzy RDD, I am a bit confused.

I think,
my running variable is year of birth
my cut off is individuals born in 1986 (normalized year of birth= year of birth - 1986)
my treatment variable is education (1=completed at least 8 years, 0=otherwise)

Under this setting, how can I estimate the causal effect of maternal education on child health? Can you please help me about the syntax?

Thank you so much.