Hi everyone,
I'm trying to match two datasets using the reclink2 user-written command on Stata 17 (note that I have the same issues when I use reclink).
My datasets contain names, states and county. I am trying to match exactly on state and county and allow for small differences in names.
I noticed that I don't get the same amount of matches depending on the variables I include, even though they are common in both datasets.
I restricted the two datasets to North Carolina only.
This command
gives me 64 matches (10 exact matches) but this command:
gives me only 10 matches (10 exact) whereas state is common in both dataset (a float = 37). Several names that are very closed to each other do not match.
Any idea what could be happening?
Alternatively, do you have suggestions of other commands I could use ? I know matchit is an alternative but it does not allow for some required variable as reclink does.
Thank you very much,
Nicholas
I'm trying to match two datasets using the reclink2 user-written command on Stata 17 (note that I have the same issues when I use reclink).
My datasets contain names, states and county. I am trying to match exactly on state and county and allow for small differences in names.
I noticed that I don't get the same amount of matches depending on the variables I include, even though they are common in both datasets.
I restricted the two datasets to North Carolina only.
This command
Code:
reclink2 county name using "NC2", idusing(id_using) idmaster(id_master) require( county) gen(similscore)
Code:
reclink2 state county name using "NC2", idusing(id_using) idmaster(id_master) require(state county) gen(similscore)
Any idea what could be happening?
Alternatively, do you have suggestions of other commands I could use ? I know matchit is an alternative but it does not allow for some required variable as reclink does.
Thank you very much,
Nicholas
