This paper shows that there are severe measurement errors regarding the occupational affiliations in the German Socio-Economic Panel. These errors are traced back to the survey structure: in years where occupational information is gathered from the entire employed population instead of only from...