I tried to upload an empty table example, but the upload fails. Any help is much appreciated.
Here is the example table.
Since the IPUMS-USA project provides microdata, to create the table you are proposing you will need to download an extract and manipulate the data in a statistical software package. You can also download the data in CSV format, which can be read into Excel. Very large files may be truncated due to Excel’s row limits. IPUMS-USA has a number of video tutorials that walk user’s through the process of creating, extracting, and opening a data file. Once the data has been read into your statistical software package or Excel, you can begin to generate your proposed table. The specific syntax you will use to create the table will depend on your chosen software.
I hope this helps. And if I can be of any further assistance, do not hesitate to elaborate on your question.
Thanks so much for your reply Joe. Here is some more info about this project:
The task of downloading and manipulating the microdata is a bit daunting and exceeds my current capacity (mostly time, but my access to analysis software is limited). Do you believe that either SDA or DataFerret might be successfully used for my purpose? I have tried to work with both, but have failed to create the table. Some advice about how to use these tools would be appreciated. Some example syntax would also be extremely helpful (if these online tools can work with syntax). One of the reasons I was unsuccessful with SDA and DataFeret was that I could not generate medians, but there were other issues too.
The weighting issue, to adjust for sampling, is also a problem that I am struggling with how to address.
I was planning to use 2013 1-year CPS data (but I would be satisfied with 3- or 5-year data if the margins of error for the single year were too high).
The variables that I believe would be ideal to use would be hhincome, race, statefip, and presence of children.
Unfortunately, I am not aware of a way to create your proposed table using SDA. SDA can produce univariate statistics, such as medians, but you would have to generate each cell as an individual table (placing hhincome in the Row: field and defining your table-cell using the Selection Filter(s): field). Furthermore, SDA does not generate margins of error for univariate statistics (you can generate margins of error for comparisons of means, as well as CIs and Standard Errors for basic tabulations, however).
I am not very well acquainted with the DataFerrett table creator, but there does seem to be a way to generate a medians table.
I hope this helps.