HHWT in the 5-year and 1-year ACS


This might be a stupid question, but I couldn’t figure it out after reading through the documentation.

When I compare the same person in the 5-year 2015-2019 ACS and in the 1-year 2019 ACS, I find different household weights. In particular, I divide the variable hhwt by 5 in the 1-year 2019 ACS and compare it with the variable hhwt in the 5-year ACS, and while they are highly correlated, they are not identical. Why is that?

Kind regards,

Hi Lorenz,

This is explained in the PUMS Accuracy of the Data report which discusses the weighting procedure for the 5-year file. The report states that the “5-year PUMS initial weight is equal to the product of the ACS 1-year final weight for the record and the PUMS subsampling factor divided by five.” You can refer to the report for more details on weighting in the ACS