Surveys covering only a random sample of companies can only be regarded as representative if the surveyed sample itself is representative of the economy as a whole.

This implies:

• Does the branch mix of the surveyed sample correspond to the branch mix of the economy as a whole?
• Do the surveyed companies adequately represent the economy as a whole as regards company size?
• Is the regional composition of the surveyed companies comparable to the regional dissemination of the companies as a whole?

The data set for Germany surveyed in 2009 exemplifies the relation between the surveyed sample and the structure of Germany’s manufacturing industry concerning sectors, firm size and region (Bundesländer).