Understanding the merge command

[Syed Raza]

Hello everyone,

I am trying to merge the data by using Pidlink which is a unique identifier in my data, hhid which is households identity and year. So when i merge two data sets the range of merge is like this




. tab _merge

_merge | Freq
------------+-------
1 | 1,774
2 | 21,719
3 | 1,254
------------+------
Total | 24,747


I want to understand what do we understand by _merge 1 , 2 and 3 ? How many pairs are uniquely identified here? Thanks

[Maarten Buis]

The meaning of the values of _merge are documented in help merge . With merge you have two files: the one currently in memory (the master) and the file you want to attach to it (the using). In your case you have 1,774 observations that appear in the master but are not present in using, 21,719 that appear in using but not in the master, and only 1,254 that appear in both files and are thus matched.