Dear David--

This is a huge problem in the ACS.  In the staqndard publicshed numbers, they use the normal approximation to compute the putative standard errro, which they turn into a margin of error.  However, for situations where the percent of given category…

Since the block populations come from the decennial census they have a 0.0 MoE.  If you assume that the population is uniform across the block then both the estimate and the MoE would scale by the same factor ( new_area/old_area ) * MoE.   I can't think…

Since the margin of error is unsigned, I would think that you should still use the sum-of-squares method when calculating aggregated MOEs. But I'm not a statistics expert.

Did you get an answer yet? Age and sex and race are controlled by estimates IIRC so there’s no margin of error unless you’re getting into more detail than the published  age-race-Hispanic-sex estimates 

Using replicate weights, in general, will result in quite low standard deviations compared to the published numbers that the Census puts out.  This is especially true for income, since income is, of course highly skewed and if one follows a reasonable…

Two things:

1 - Tim is right. Total population for a county is an estimate from PopEst, and doesn't have sampling error, so margin of (statistical) error is 0.

2 - Formula for calculating aggregated MOE is the same whether you're summing or subtracting…

I would love to know the answer to this.