What is variable to identify IPUMS CPS Basic March cases from supplement cases in a given year IPUMS CPS ASEC?


We plan to make crosstab for % married by age by race for non-Hispanic white and non-Hispanic black from year 1988 to 2013 and separate by gender. IPUMS CPS Basic March is our original intended data source. But there is no IPUMS CPS Basic March for 1988, and it is important for our research to start the year in 1988. We try to use ASEC samples and select cases that are from basic March and delete the additional oversample/supplement cases. However, we cannot find information to identify basic March cases from supplement cases in ASEC samples.

We had extracted our research interested variables and auto pre-generated variables for IPUMS CPS Basic March samples for all the available years without any case selection and had another extraction for IPUMS CPS ASEC samples for the years 1988 to 2014 without any cases selection.

Would anyone please shed a light and point the direction for us?

Thank you very much,



Identifying the March Basic respondents within the ASEC samples is complicated by the fact that the ASEC includes an oversample of households from adjacent months. IPUMS-CPS is currently working on a variable called MARBASECID to facilitate linking the two data sets. The attached article goes into detail about the development of MARBASECID and the difficulties associated with linking the ASEC to March Basic responses. As noted in the pdf document, this article is still in draft form and should not be cited.

Also keep in mind, some variables have different universes and reference periods depending on whether they appear in the ASEC questionnaire or March Basic questionnaire. Therefore, the 1988 ASEC sample may not be a perfect substitute for the 1988 Basic sample. I mention this as a warning, not to discourage you from attempting to use the 1988 ASEC data.

I hope this helps!

PacasFlood_06-05-15-2.pdf (517 KB)