HHWT or PERWT for Derived Families

I am doing a parity analysis, so I’m looking at counts of newborn babies, divided up depending on previous births to that mother (in the household). For example, how many babies were born to a mother that already has one son? one daughter? A son then a daughter? etc.

So while I am counting the individuals, they’ve been formed into family units that might be more similar to households.

I am using the 2007-2011 ACS 1-year data.

In this case should I be using the baby’s weight, the mother’s or the households?

Since there is no specific weight associated with families, it primarily up to the researcher to decided which weight to use. According to Chapter 11 of the ACS Methodology page, the Census Bureau uses Housing Unit Weights for family-level estimates. I would recommend looking into other family-based research using ACS data to see if there is a common approach in your field.

I hope this helps.