I have used ACS data quite a bit, but recently a question came up that I can't find the answer to. Are non-primary residences included in the data? The instructions say to include people who are not away for more than 2 months, but it doesn't seem to even specify whether that is per year or a 2 months continuous period. What I am most interested in is people with 2nd homes - their primary residence is elsewhere. Are they included in the data?
This question has come up because I have often compared data from the IRS and from the ACS. Usually, they show similar comparisons of areas (given the different measures that they use). The IRS data is linked to the address on the tax return. The ACS data is linked to the address the survey was sent to. I have come across a case where an area I am looking at has relatively high income from the IRS data but relatively low income in the ACS data. I am trying to figure out if this is due to the presence of a large number of 2nd homes in the area (of which there are many).
Can anyone help clarify the extent to which non-primary residences would be included?
Thanks in advance.
After a bit more research, it appears to me that 2nd homes would generally be reported as "vacant." What I can't determine is whether income is reported for vacant units or not. If it is a vacation home and thereby put in the vacant category, do respondents answer the income questions and are these reported in the data?
2nd homes are generally classified as vacant "for season, recreational, or occasional use". And no income would be reported for those. Income goes with residents - so it would only be recorded for what you consider your usual residence and not for a 2nd home. if it is vacant, it is vacant so no household information (people records) would apply.
oh.. check out table B25004 and B25005 if you want to try and get numbers on 2nd homes.
Jamie Rae gave you a good place to look for places listed for 'occasional' use though it's hard to know how they many they would miss and I don't know how anybody could ascribe income to them ... I'll tell you a story about Miami-Dade County, FL after 2000: there's a municipality made up entirely of gated mansions used occasionally by wealthy owners and it ended up getting a fairly high poverty rate in that Census. I tried to figure out why. Did people have gardeners or housekeepers living on site? Impossible, said the mayor. I know everybody here and there's no poverty, she told me. Back then people in the Census Bureau would actually talk to you about things and someone did get back to me and say this: Since they hadn't been able to reach anyone at most of these places, they filled in a lot of data from a nearby area of similar size across the bay in what was then impoverished Northeast Miami.
Thanks to you (and Tim) for the quick responses.
Be cautious in any assumptions. ACS survey questionnaire includes instruction to complete or include people "living or staying here for more than 2 months." And so, if the survey shows up during "vacation season", when the seasonal 2nd home is occupied, and people fill it out... maybe that's a false positive -- but it's difficult to know.A 2nd reason for caution: the "residence rule" for ACS survey is completely different than residence rule for filing taxes. So if your geography of study includes tax-avoidance havens (like Florida), it's very possible there will be more tax returns from some places than there are year-round residing households.