[Rajdeep Chaudhuri]

Hi all,
I was just wondering if matchit allows a many-to-one matching. For example, if I have many entries of a name in one dataset and then only one entries of the same name in the other dataset, would a match be possible using those datasets?

[Clyde Schechter]

I'm not sure I understand your question. -matchit- does many-to-many matching. It matches each observation in the master data set to as many observations in the using data set as are plausible matches according to your option choices. If it turns out there is only one plausible match for some observation, it returns only that plausible match for it.

[Julio Raffo]

Clyde Schechter is (as usual ) correct. -matchit- does many-to-many (m:m), which allows for m:1 and 1:m particular cases. However, depending on the two datasets sizes, I always recommend running unique names (i.e. unique instances of str variable) against unique names to avoid running redundant comparisons.