Ensuring Treated Subjects are Matched to Control Subjects

Question

The entry for ratio in Matchit::matchit reads

for methods that allow it, how many control units should be matched to each treated unit in k:1 matching. Should be a single integer value. See the individual methods pages for information on whether and how this argument is used. The default is 1 for 1:1 matching.

In my study, I plan to estimate the ATC (average treatment effect in controls) since we're interested in knowing the effect of the treatment were we to give it to people who would not typically get the treatment. We want to use matching as one form of estimating the treatment effect, and hence we want to match multiple treated individuals to controls

I wanted to clarify that if I specified estimand = "ATC", and I set ratio=3 then I would be matching 3 treated individuals for every control. The documentation makes me think matchit might be matching 3 controls to every treated individual, regardless of the estimand argument.

Answer 1

Yes, if you set estimand = "ATC" and ratio = 3, you will match 3 treated units for each control unit.