SDA computing variable for size adjusted HHINCOME

Hi there! I’m trying to calculate medians of household income for RACED groups, adjusted for household size, using the SDA online data analysis tool.

I’m planning on doing an equivalence scale adjustment to HHINCOME, by diving household income by the square root of the number of people in the household. To do this, what I’m doing is creating a new variable: HHIADJ = HHINCOME / sqrt(NUMPREC). I’m setting the Missing-data code for the new computed var as 9999999.

I’m a little confused by the documentation in SDA on variable computation and missing data codes, but my understanding is that all instances of HHINCOME = 9999999 are now stored in the new variable with the missing data code, so HHIADJ = 9999999.

If I go to calculate median HHIADJ under the Means tab, are the missing data automatically excluded from the median calculations? I tried to test this by filtering for HHIADJ(0-9999998) and got slightly different results. Is it because of instances of HHIADJ between 9999998 and 9999999, or am I doing something wrong?

I’m sure I must be missing some elementary information about how missing data codes work in SDA, but I’ve looked high and low and still feel paranoid that I’m doing something wrong.

Thanks in advance for any help!

Your thinking on this is mostly correct. However, HHIADJ will only equal to the missing variable code when HHINCOME is missing and NUMPREC is equal to one. My recommendation is to, after creating HHIADJ as specified, input HHINCOME(*-9999998) into the selection filters on the means tab. This will exclude all cases where HHINCOME is missing while also including all negative values of HHINCOME. I suspect the reason that you get slightly different results with your filtering is because you are excluding negative income at the bottom by setting the filter to start at 0. HHINCOME is the sum of the income of all household members taken from INCTOT. You’ll notice that that there are differing bottom codes depending on the sample in question, but they tend to be at negative levels of income. When I did the analysis using the SDA tool for ACS 2001-2019, I got a bottom value of $39,996 for HHINCOME. Additionally, there are no values of HHINCOME between 9999998 and 9999999 since the variable only consists of integers.

Hope that helps!

Yes, that helps—thank you for the clarification!