Hello: I am interested in identifying individual level data for migration from one state to another state, from one year to the next, for the USA. So in order to uniquely identify individuals, how do I proceed? From reading the online variables codes, the combination of PERNUM, DATANUM, YEAR and SERiAL shoul identify the person uniquely. I would like to combine that with MIGPLAC1 to find out individual level data for migration from one state to the next, on yearly basis. Is the right way to do it? Thank you for all the assistance you can provide.
That is correct. YEAR, DATANUM, SERIAL, and PERNUM together uniquely identify an individual within a given sample. YEAR and DATANUM identify the person’s sample, SERIAL identifies the person’s household within the sample, and PERNUM identifies the person’s location within the household. You can then use MIGPLAC1 and STATEFIP to determine the migration flow, if any, for each individual in your dataset.
I hope this helps.
Hi, a question of clarification. The combination of PERNUM, DATANUM, YEAR and SERiAL uniquely identifies the person in the IPUMS-USA samples only? It is correct that it would be PERNUM, YEAR and SERiAL to uniquely identifies the person using IPUMS-CPS?
Thank you.
The combination of YEAR, MONTH, SERIAL, and PERNUM uniquely identifies a CPS respondent within a sample. YEAR and MONTH identify the person’s sample, SERIAL identifies the person’s household within the sample, and PERNUM identifies the person’s location within the household.
Unlike with IPUMS-USA, respondents in the CPS appear in multiple samples. Specifically, a household is in the CPS sample for four months, out for eight months, and then back in for four months. The variable CPSIDP can be used to identify respondents across samples. Please refer to this previous answer for more information about using CPSIDP.
Hope this helps.