In order to use some variables in the NBER CPS monthly files, I am currently working on linking NBER CPS monthly to IPUMS CPS monthly data files by using (1) “hrhhid”, (2) “hrhhid2”, (3) “lineno”, (4) “year”, and (5) “month”. After linking two data sets, I found that there are “not matched” observations in both NBER and IPUMS CPS monthly files. The all “not mathced” observations in the NBER CPS monthly have pulineno=-1. But what I don’t understand is “not matched” observations in the IPUMS CPS monthly.
I read the previous question and answer in the “Why are there differences in the number of observations between IPUMS CPS and NBER data files?”
The answer for that question was (1) oversample in IPUMS CPS and (2) non-response households in NBER CPS monthly. According to this answer, the “not matched” observations in IPUMS CPS should be oversample. But I still have concern because I downloaded only IPUMS CPS monthly samples. Is there a way to check whether “not matched” observations in IPUMS CPS monthly are over sample? Or is there any other reason for these “not matched” observations in IPUMS CPS monthly files?