I know that the IPUMS-USA data extract system allows for this output, so that I can study vacant household characteristics such as rent or home value. But I would like to use the SDA if at all possible, as my access and training in statistical software is limited. Thanks.
While it is possible to perform household-level analysis (by using pernum(1) as a filter to select only one person from the household, and switching the the household-level weight), because the online data analysis tool uses the rectangularized version of the data, there are no vacant households in the online analysis tool (as vacant households do not have people to be attached to).
I’m sorry I don’t have better news.
Thanks for the answer, Joe. I’ve experimented already by downloading household-only data (e.g. Cincinnati households by vacancy status, hh income, home value, tenure) via the iPUMS-USA extract system in .csv format. Do you think it is reasonable to attempt to use this data in Excel, since a weight is included with each household record? I don’t want to present data incorrectly, or in a statistically insignificant manner, but for simple percentages I’m hopeful that Excel manipulation will suffice. Otherwise, I may need to look into SAS or SPSS. Thanks again!
The primary obsticals for using microdata in Microsoft Excel are the column and row limit. If you are able to fit your needed data within these limits Excel is a fine tool for generating weighted percentages. As far as I know, when it comes to performing regressions or generating robust standard errors Excel is less equipt, but there may be options for those types of computations as well.
Good to know. My target geography is fairly small and so my data will be limited to just a few thousand lines, not a problem for Excel. I’ll see how it goes. Thanks again for your insight!