Number of households

Dear team support,

I am using phase VII survey for samples, Ghana, Malawi, Rwanda, Tanzania, Uganda and Zimbabwe, collectively with children coded file for women and children participants and study variables. Is there a STATA command to identify the total number of households with each group of participants?

Thank you for your response to my last enquiry on BREASTMO. It was helpful to get that clarity during this re-analysis stage.

Looking forward to your expert view.


SERIAL is an identifying number unique to each household in a given sample. All person records are assigned the same serial number as the household record that they follow. The combination of SAMPLE and SERIAL provides a unique identifier for every household in the IPUMS-International database.

If you are using a single sample, the simplest way to do this in Stata is to use the -codebook- command to find the number of unique values of SERIAL.

If you are using multiple samples, you need to find the number of unique combinations of SAMPLE and SERIAL. One way to do this is with the user-written command -distinct-. Install by typing net install dm0042_1. Then type:

distinct sample serial , joint