I am doing a project calculating economic sufficiency based on total income along 6 axis: married couples that own their own homes with a mortgage, married couples that own their own home without a mortgage, singles that own their own home with a mortgage, singles that own their own home without a mortgage, married couples that rent a one bedroom, singles that rent a one bedroom.
I am using ACS data from the years 2009-2013.
I selected for state (CA), county, and age (over 65). I attached a characteristic to age- head of household and also to INCTOT- head of household and spouse.
I created 6 new variables based on the axes above, including in each one that the head of household has to be 65+.
I want to sort by county, tabulating each new variable and summarizing the income in each case.
In stata I use the command for each variable that involves a single person: by county: tab (variable name), summ (inctot_head).
I do I caclulate the right income in each case for the married couples? Should I use HHINCOME, FTOTINC, or run inctot_head and inctot_spouse separately for each variable and add the totals?
Using HHINCOME or FTOTINC will be too inclusive for your purposes. HHINCOME sums the income for all persons in the household, and FTOTINC sums the income for all persons in the family unit. You are correct that you will need to instead sum the head of household income with the spouse income.
Also, keep in mind that it is possible to create a variable within the IPUMS data extract system that contains the income of a respondent’s spouse by utilizing the Attach Characteristics option. To do so, you should first select your samples and variables, which must contain INCTOT. Before submitting your extract, you will be given the option to choose “Attach characteristics” on the extract request screen. Check the box for “Spouse” on the INCTOT row. This will add a variable to your data extract request called INCTOT_SP. Now simply submit your extract. You should then add up inctot and inctot_sp for one spouse member.
In calculating the average total income for the head of household I use HHWT and PERNUM=1 with INCTOT. Can I simply add INCTOT_SP, with PERNUM=1 still applied - i.e. are the spouse numbers relative to PERNUM 1?
If you intend to calculate the total income of spouses and household heads, your approach is correct. Note that this approach will be excluding some couples from your analysis, because by restricting your analysis to only individuals with PERNUM==1 (e.g. household heads) you are restricting your calculations to only including at most one couple per household. In reality, some households could have couples of which neither of whom are the household head. This may not be a big concern to you since you seem to be interested in the average income of the household head and their spouse.