I am extracting a dataset that is primarily at the person level, but I only want to look at one month-in-sample rather than having households appear multiple times in the data. Selecting cases by MISH would greatly reduce the size of my data extract, which is quite large since it comprises all basic data since 1994. Why isn’t MISH listed as an option for selecting cases? I realize that it’s a household variable while I am using people, but it’s a variable that is completely consistent over the years and there are numerous reasons why a researcher might want to only look at samples at certain points in their rotation. Thanks.
This is a very good suggestion and I will pass this along to the IPUMS CPS team to consider for their next data release. In the mean time, a more tedious solution would be to break up your extracts into smaller chunks of samples. You can then select the MISH value from within your statistical software, dropping all other cases, and save the subsetted files. Since these sets of samples will include all of the same variables you should then be able to append them all together resulting in your desired, single MISH data file.
I hope this helps.