Hello, I am looking to analyse the change (if any) in the political opinion of an individual after they transition out of a marriage. I have a panel data set and would like to carry out propensity score matching followed by a multinomial logistic regression. I have a variable treatment=1 for the specific year an individual transitions out of a marriage and treatment=0 for observations of individuals who are continuously married.

Would I be able to analyse the change in my outcome (political opinion) by simply defining my treatment as above? How can I measure my baseline covariates and analyse if they change a year after their transition out of a marriage?

Many thanks,
Sarah