The family income data in ATUS is in nominal terms. So, if we merge several rounds of ATUS data should we deflate the family income value using Consumer Price index?
Since family income data is given in categories (ie ranges) how to deflate?
You are correct that it’s necessary to adjust incomes to a given base year. You can find the adjustment factors on this page. Adjusting categorical income however is tricky and is up to the researcher’s discretion. One method is to assign a single value to the interval, such as the midpoint, and adjust all observations within this range for inflation using that value. I also recommend taking a look at what methods others have employed in the literature; a google search provides a different suggestion here.
Alternatively, you may be able to derive the exact family income by linking respondents to their final CPS interview where they responded to a set of outgoing rotation group questions including EARNWEEK and WKSWORKORG. I’m happy to provide more information on how to go about linking respondents and reconstructing family income if that’s something you’re interested in.
Dear Ivan
Thank you very much for your informative email.
I had used CPSID to get actual family income and then deflated it using the annual deflators. So that’s done with for the time being !
However, I do plan to work with ATUS data. So your email and the other methods that you suggest will be useful. If you do not mind, I will get in touch with you again when I resume work on this area later this summer.
Best regards
Zakir