Merge - label already defined

[Friederike Osterndorff]

Hi,

I am new to stata and I have to merge various panel data sets from different waves. The good think is they are all from the same source, so the identifier is given and the same across all waves.

For the first three datasets the merge went fine, however, now stata gives the warning xxx label already defined. What does that mean and how can I fix it?

Also the number of not matched observations is going up, with every merge. Is that a problem?

Thanks.

Best,
Friederike

[Rich Goldstein]

re: label already defined - this just means that in both the master and the using data there is a label defined with the same labelname - you need to check whether the definitions are the same in the two files; if not, an "easy" answer is to drop the labels and then redefine after all merges are completed

in multiple wave data it is often the case that people drop out (and sometimes new people enter); so this may be ok but you need to check your data carefully to be sure that some error has not entered; you don't say how many "not matched" observations you have each time so it is hard to assess the seriousness of this

[Friederike Osterndorff]

Hi, thanks for the answer. It was really helpful. I will check the data to see what is up with the labels.

Also I found out that I was not merging the data correctly. I changed my approach and now I have 0 unmatched observations, but thanks.