March 2007 CPS incwage data has many recurring values (i.e. "619221") without any correlation b/w observations.Why

#1

I am looking at the March 2007 CPS “incwage” data, and have noticed several exactly recurring values (i.e. “619221”, “579599”, “446853”) for totally uncorrelated observations (i.e. individuals from different states, ages, etc.) - what can explain these recurring values? Are they inoccuous or am I missing something?

0 Likes

#2

IPUMS-CPS does not transform or adjust income figures, but directly adopts the values in the original PUMS files provided by the Census Bureau. I can not say for sure why such strange reoccurring values would exist, but they do occur in every ASEC sample (though at different values). My best guess is that it has to do with the Census Bureau’s Disclosure Avoidance measures. The definitive source, however, would be the Census Bureau, as the values originated in their PUMS files.

0 Likes