ACS Data Users Group

A colleague was recalculating derived variables from the ACS for a research study and is trying to calculate Margin of Error for population density using guidance here: https://www.census.gov/content/dam/Census/library/publications/2018/acs/acs_general_handbook_2018_ch08.pdf

Should he just assume that the MOE for land area is 0 and use the formula presented there (p.56)? Or is there a different approach he should take for when only the numerator has a published margin of error?

Hi Rebecca,

Here's what I did recently in collaboration with Census for their COVID-19 Hub site, basically treated the land area value as a constant:

B01001_calc_PopDensity = 1,000,000 * B01001_001E / ALAND

MOE(B01001_calc_PopDensity) = 1,000,000 * B01001_001M / ALAND