Propensity Score Matching using R
It is a statistical matching technique that attempts to estimate the effect of a treatment, policy, or other intervention by accounting for the covariates that predict receiving the treatment.
Try to eliminate the effect of "Treatment" and check if other variables have the same effect on outcome or not. Here, it's for a music app "High Note". The outcome is the likelihood of a customer being an 'adopter', i.e. a paying user. The treatment is having friends that are adopters. We check the effect of other variables in treatment and control group, and also after removal of treatment.