Constructing MAGI in IPUMS-CPS


Can I use the dataset to construct the modified adjusted gross income (MAGI) for each respondent? If so, could you please briefly run down the steps.




IPUMS-CPS does provide a number of Tax Variables from the March Supplement, including ADJGINC which gives the respondent’s adjusted gross income. However, these tax variables are actually calculated by the Census Bureau based on multiple data sources (as discussed briefly in the ADJGINC variables description, this IRS publication provides more information on how the figures are calculated), rather than based on respondent answers to direct questions. So, unfortunately, IPUMS-CPS does not have data on things such as IRA contributions, student loan interest, or the like. I am sorry I could not provide you with the data you were looking for.



Thanks for the response, Joe.

I have a follow-up questions. Are there any variables in IPUMS-CPS that are syonomous with gross income and net income?



The variable INCTOT can be considered the Gross Personal Income figure. You can then use the Tax Variables to figure their net income. Looking further into these variables I see that the variable TAXINC may come close to a MAGI measure, as it represents Adjusted Gross Income minus “imputed itemized deductions or the standard deduction”.

I hope this helps.