I'm using matchit package to create a propensity match. I'm trying to match control and treated with a 2:1 ratio in order to maximize the population and exclude less patients as possible (13.000 patients available). I have found two possible ways and I am wondering which one is better:

  1. method=nearest, ratio 2:1, min.control=1 and max.control=3. In this way I obtain a good matched population with precise 2:1 ratio and good sample size (almost 12.000 patients)
  2. method=nearest, caliper = 0.1 and a ratio of 3:1. In this way I still obtain a good match but a not completely balanced population. Are the two ways good options that avoid risk of bias or one of the 2 is superior to the other one?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.