What observations would be coded as observations not applicable in MIGRATE1?
The Universe Tab for MIGRATE1 states that applicable respondents are all those who are age over 1 years old. (This is also true if you are using MIGRATE1 in IPUMS CPS.) Therefore, in the data you’ll see that all those who have MIGRATE1==0 also have AGE==“Less than 1 year old”.
I find that there are 54 observations in 2002 where MIGRATE1 = 0 but age is not “less than one year old” (this is not true of any other year, 1981-2018). Do you know why this is the case? Thanks.
Thanks for the note about this issue. I have replicated your observation, there are definitely 54 observations in MIGRATE1 who are one year old or older and who are NIU in the 2002 ASEC sample. I’ve discussed this issue with the IPUMS CPS Team and it looks like these observations are coded this way in the input data we received from the Census Bureau. Therefore, there is nothing much we can do to the data but we will post a note about this issue on the variable description.