Regression Discontinuity

Explanation

It is a method that looks at the discontinuities in regression lines at the point where the intervention takes place. The idea here is that, in the absence of the causal effect, the treatment effect should be very similar on different sides of a sharp cut-off point. This holds true if we can assume that the population which is very close to the cut-off point are similar with respect to any relevant confounding variables. If this assumption holds and there is a clear discontinuity in the causal effect at the cut-off point, this is evidence that crossing the treatment or the threshold point has a causal impact on the population.