ACS special tabulations

As a follow up, here is what I might be requesting. Anyone know whether such a request is possible? I was looking at one table, s1702, poverty. They have poverty by race, but only White has data for non-Hispanic. Is it possible to to make a special request for Black non-Hispanic, Asian non-Hispanic?

As a follow up, here is what I might be requesting. Anyone know whether such a request is possible? I was looking at one table, s1702, poverty. They have poverty by race, but only White has data for non-Hispanic. Is it possible to to make a special request for Black non-Hispanic, Asian non-Hispanic?

Can you use PUMS?

Hi Mara. Would I be able to get data for every county in New York State? By the race/ethnic groups I was looking for?

Only if the counties have more than 100,000 people. Here is a list of NYS PUMAs:

https://www2.census.gov/geo/maps/DC2020/PUMA/st36_ny/

You can also browse them online here:

https://tigerweb.geo.census.gov/tigerweb/

Okay, thanks. A number of counties in New York State have total populations less than 100,000, so I guess not for the PUMS. Thanks though.

I did some "back and forth" with the person who is responsible for ACS special tabulations.   It is nice that the landing page mentions the $3000.00 minimum charge. When I looked back in April I had to guess at what it might cost.  The following document (2015 but I think that it might be current) is helpful. https://www2.census.gov/foia/ds_policies/ds021.pdf    You will have to write a detailed memo giving a specification for the table including variables and the geographies.  It is useful if you have experience writing a description that could work as programming specification for a SAS programmer. The PUMS documentation is good for seeing what variables might be available. Remember the PUMS variables are restricted because the PUMS geography goes down to about 100,000 persons. Once you have the proposal worked out (no charge for this step in the process) it we be presented to the disclosure (avoidance) committee and they will give you a yes or no.   What you can do depends not only on the variables but the geography. The larger the geography the more variables you can use in the tabulation.  I asked about a "PUMS" like file, i.e. a sample of records (suitably dithered) for a small set of variables.  The response I got was a definite "no." It took a while for them to get back to me so the decision  required some internal discussion. I think that the basic expense is the programming hours to program and run the request.  I did get some good news -- no extra charge for additional geographies.  Also don't forget to check the existing special tabulations (there is a web page listing them). You might get lucky and there may be an existing special tabulation that will fit your needs.  I think that there will be no charge in that case.  Legally this comes under an FOIA (Freedom of Information Request) regulation (see document link earlier).

Good luck and get out your wallet !

Dave

PS think about a multivariate tabulation with extra variables at a larger geography then use the "regular" ACS tables at the lower geography together with Small Area Estimation go get what you want for a geography smaller than the custom tabulation.

I think it would benifit everyone if a group such as this or in the upcoming meeting one or two new tables were submitted (sort of what the group would really like to see) and they were not overly complex. Free of charge of course.