Denne siden på norsk

Control of anonymity

  1. Do the data contain directly identifiable personal data such as personal identification numbers, names, addresses or phone numbers?
  2. Do the data contain indirectly identifiable personal data? This is determined by:
    • combining different information, e.g. geographical information/variables (municipality number, ward or district) with other background information such as income, education, occupation or position.
    • checking the sample size. In a small sample, information such as gender or age can make the personal data indirectly identifiable.
    • checking the composition of the sample. If, for example, a sample's gender or age distribution is biased, it may be enough to make the data indirectly identifiable.
    • checking the data's degree of detail. If, for example, the occupation and education variables are very detailed (four or five-digit level), the data set is more likely to be indirectly identifiable.

The result of this control decides the further processing and storage. NSD distinguishes between anonymous and personal data.