Accuracy and reliability

Overall accuracy

The general quality and reliability of the Civil Registration System is regarded as very high. However, looking at relation one has to take caution for certain groups in the population.

Older generations

When CPR was established in 1968 only people alive at the time was given a CPR-number. Thereby the number of parents, grandparents and siblings (since siblings are found through shared parents) for the older generations in the population are underestimated. Further CPR was intended for tax purposes in the early years of CPR, at the time every family was a tax-unit. When a child moved away from home or turned 18 they were considered their own tax-unit and the link between parent and child were annulled. This results in persons born before 1967 the number of siblings and parents are underestimated. For persons born before 1935 the number of parents, siblings, children, children-in-law and grandchildren may be very underestimated, and caution is warranted looking at the mentioned types of relations for these age groups. Attempts are made to correct the missing parent-child relations and this has improved the number of relations created.

Immigrants

All types of relatives, not established by shared home address or children born in Denmark, are underestimated for immigrants. This is due to lack of knowledge of family members who have never been a part of the danish population. It is also du to lack of linking parent child at immigration to Denmark. The latter is to some extend corrected over time.

Connecting couples

When connecting couples it is assumed the two people are a couple when they are mixed-gendered, sharing an address and the age gap is less than 15 years. It has not been checked how many of these couples living together are actually a couple. Some will probably be eg. two students sharing an apartment.

Two same-gendered persons, without shared children, will not be counted as a couple. Hence, the number of same-gender couples are underestimated.

Some comparing have been made on self reported couples in questionnaires and and those couples identified with Statistics Denmarks e-familytype. These comparisons shows that for statistical purposes, the register definition is representative.

Newborns

There will be some new-borns that not yet has a parent 2 in CPR at the time of reference, as there can be up to half a year delay in this registration. This is hence especially a problem with newborns born in the second half of the year before counting, and for newborns where the parents are not married at the time of births. If one is interested in the relatives of newborns, one should not look at the latest year of reference.

Parent-child relations

Parent child relations in CPR is an administrative relations, which cannot necessarily be sided be legal or biological parent-child relations. As no systematic checks are made on parent-child relations in CPR, cases can be made were two persons have more than one type of relation. The number of these cases are however limited.

Sampling error

Not relevant for these statistics.

Non-sampling error

Incorrect addresses, which affect the assessment of families and households in particular, involve certain challenges. People without a permanent residence are a source of uncertainty. They are registered under special municipal addresses of a fictitious nature and are managed by the local civil registers. These addresses are reserved for people who cannot be registered at a residence for some reason. E.g. homeless persons, sailors, posted diplomats and certain other posted persons. Read more about the quality of the municipalities address registration in a survey by VIVE (in Danish). For people living on incorrect addresses the distance to their relatives cannot be calculated.

Another source of uncertainty regarding households is communal dwellings (e.g. nursing homes) where the residents are sometimes registered under individual addresses and sometimes with a common address.

As a result of both CPR and shared home address are used in creating relations, cases can be seen where two persons have multiple types of relation. The amount is however limited to about 500 unique relations between any two persons a year.

Quality management

Statistics Denmark follows the recommendations on organisation and management of quality given in the Code of Practice for European Statistics (CoP) and the implementation guidelines given in the Quality Assurance Framework of the European Statistical System (QAF). A Working Group on Quality and a central quality assurance function have been established to continuously carry through control of products and processes.

Quality assurance

Statistics Denmark follows the principles in the Code of Practice for European Statistics (CoP) and uses the Quality Assurance Framework of the European Statistical System (QAF) for the implementation of the principles. This involves continuous decentralized and central control of products and processes based on documentation following international standards. The central quality assurance function reports to the Working Group on Quality. Reports include suggestions for improvement that are assessed, decided and subsequently implemented.

Quality assessment

These statistics are based on administrative data sources with full coverage. Undercoverage has been found for certain groups in the population. These statistics are timely, as it covers relative relations that, has not previously been made publicly accessible. In general these statistics are expected to meet user demands. It is however the expectation that there will be a demand for further measurements.

Data revision - policy

Statistics Denmark revises published figures in accordance with the Revision Policy for Statistics Denmark. The common procedures and principles of the Revision Policy are for some statistics supplemented by a specific revision practice.

Data revision practice

A revision has been made for the years 2019-2022 and 2024 in a version 2 of the register. The revisions included the distance in birds-eye-view between two home addresses and was published in march 2026. The revision did not affect already published statistics.

In the future a version 2 of the latest year of measurement is expected to be published. Version 2 will include the distance between homeadresses measured in roadmap. The exact method for this revision is still under development.