This is a very interesting table note to an ACS table.
This table, of sex by age
https://data.census.gov/cedsci/table?t=Populations%20and%20People&g=0400000US36&d=ACS%205-Year%20Estimates%20Detailed%20Tables&tid=ACSDT5Y2018.B01001&hidePreview=true
has **** in the margin of error column, and this note of explanation
“An "*****" entry in the margin of error column indicates that the estimate is controlled. A statistical test for sampling variability is not appropriate.”
I’m not really sure what this means. Any help would be greatly appreciated!
Thanks
Gene
Essentially it means it has a SE of zero. Because the weights are designed in such a way that the certain margins (age, sex, et al) are fixed to independent estimates, usually from the population estimates…
yes as Matthew mentions this comes from the annual Age-Sex-Race-Hispanic estimates so not from the survey which would have sampling error.
I would just think it as one of the variables used to factor and adjust the sample. This is not the statistical answer but its the way I came to peace with the footnote.
Essentially it means it has a SE of zero. Because the weights are designed in such a way that the certain margins (age, sex, et al) are fixed to independent estimates, usually from the population estimates program, the estimates from ACS are treated as if they do not have sampling error.
Thanks all. I think this explains it, "Because the weights are designed in such a way that the certain margins (age, sex, et al) are fixed to independent estimates, usually from the population estimates program, the estimates from ACS are treated as if they do not have sampling error."