ACS Data Users Group

Hello, I'm working on adjusting median household and family income at different levels of geographies using ACS 1-year and 5-years estimates. For example, I want to adjust 2005 ACS 1-year median family income estimates to 2014 dollars. I can do that using the ACS Handbook guide but how do I deal with the margin of error for the new 2005 adjusted estimate?

Thanks in advance!
Gustavo.

Hi Gustavo,

First I want to be clear that I'm no expert in this sort of thing (I'm a computer programmer, not a population/statistics guy), but from what I've seen, the margin of error applies to household & people counts, not to incomes. What you are doing wouldn't affect person/household counts, so if I understand this correctly (and please feel free to correct me otherwise!), margin of error would be handled the same as if you hadn't done the adjustments.

- John Grumbine

The units for the estimate and for the margin of error are the same. For median income, I THINK (please correct this info if it is wrong) that the units are in inflation-adjusted dollars for the ending year of the ACS time period - for example, the 2005-2009 median incomes are in 2009 inflation-adjusted dollars.

Any adjustment to 2014 dollars that is made for the median income estimates should also be applied to the margins of error.

After reading David's response, I realized that the answer may depend on whether you're using PUMS or pre-tabulated data.