Merging datasets & matching unique IDs

[Vanessa Garcia]

Hello! I have two datasets that I would like to merge using the clients' unique IDs. However, only one of the datasets has the unique ID variable and the other one does not. I do have names and dates of birth for both datasets and wanted to know how I could merge these datasets and match the clients with their unique IDs and/or names and dates of birth. There are a few clients who have the same first full name and last name initial but different dates of birth and wanted to account for this as well. I would deeply appreciate any help! Thank you in advance!

[Leonardo Guizzetti]

This might be clarified by a data example, but if I understand you correctly that there is only the client ID in one dataset and not the other, there is no way to link these two datasets.

[Vanessa Garcia]

Thank you Leonardo, so I won't be able to link the datasets even if the dataset without the client ID has the names and dates of birth of the clients?

[Leonardo Guizzetti]

Oh I had misunderstood your first post. As long as both datasets have the client name and date of birth, and you are sure that you don’t have clients with identical names and dates of birth, then yes you can link them. These three variables become your merge key. See either -help merge- or -help frlink- for applicable commands.

[George Ford]

Look for duplicates by name/age in the merging dataset.

[Vanessa Garcia]

Got it! Thank you so much!