Why do MIGMET1 and METAREA labels differ? And what precisely are the N/As in MIGMET1?

I have noticed that the names of certain MSAs differ between METAREA and MIGMET1. Some examples include “Chicago, IL” vs. “Chicago-Gary-Lake, IL” and “Boston, MA/NH” vs. “Boston, MA”. From the variable descriptions it seems that there should be no difference between the labels of these two variables, certainly for ACS samples through 2011. Can I assume that they refer to the same MSAs?

Second, I asked a follow-up question here about the N/A code in MIGMET1, but I am not sure if that was the right place to ask. There it was said that MIGMET1==0 “N/A” identifies people who were outside of the question universe. However, there are around 4000 respondents each year for which the following condition is true: currently living in an identifiable METAREA AND MIGMET1 == “N/A” AND MIGRATE1 == “moved within state, within PUMA”. Do MSA boundaries frequently cross PUMAs?

Thank you for your help.

It looks like the discrepancy is being generated by the difference between “general codes” and “detailed codes”. The METAREA variable has both types and the general code represents the first three numbers in the detailed code. The detailed code includes more specific information about the metropolitan area. These detailed codes seem to align better with the MIGMET1 variable codes. More information on general codes vs. detailed codes can be found here.