INCTOT for 1950



According to the Comparability Tab under INCTOT (Total personal income), it is indicated that INCTOT (total personal income) is the sum of INCWAGE (wage and salary income), INCBUSFM (business and farm income), and INCOTHER (other income). (…) However, as you can see:

On the fourth line, INCTOT = 350, INCWAGE = 0, INCBUSFM = 250, and INCOTHER = 150, which doesn’t make sense as 350 ≠ 250 + 150.

There were several similar cases, which caused issues when I was calculating ratios such as (INCWAGE/INCTOT, INCBUSFM/INCTOT, INCOTHER/INCTOT), and the ratios exceeded 1 (or 100%).

Is this just the technical issue? Or am I understanding something wrong?

Thanks for your help.


This is largely a technical issue. The codes for INCWAGE, INCBUSFM, and INCOTHER all represent midpoints of hundred-dollar intervals. So, do the codes for INCTOT. The discrepancies you note, are consequences of this fact. The questionnaire in 1950 asked for a specific number, the intervaled codes were added later by the Census Bureau. So, although the true sum of INCWAGE, INCBUSFM, and INCOTHER equals what is then intervaled and represented by a midpoint via INCTOT, summing these variables as they are presented in the public use data won’t always replicate the true sum.

I hope this helps.