Accuracy and reliability

Overall accuracy

One source of error which is difficult to detect is when a municipality makes the same administrative error year by year. The municipalities might overestimate or underestimate certain variables because they are uncertain about how to estimate different services. It is not possible to establish whether the variables are consistently under- or overestimated. A certain statistical uncertainty is linked to the different administrative practices in municipalities.

Sampling error

These statistics are based on data from all municipalities. Consequently there is no sampling error.

Non-sampling error

The survey is a full-scale census based on responses from all municipalities and all questionnaires are usually received. In some cases data from the previous year are used in agreement with the municipality. In some cases it is difficult for the municipalities to provide the correct data from their systems. In these cases data received might be less accurate. Further more some municipalities change their way of reporting data to Statistics Denmark or realize that previous data have not been reported correctly. Further more change of systems in the municipalities might cause changes in the reported data. This might cause fluctuations in data over time for certain municipalities.

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

The quality of data is improved by contacting municipalities that report large deviation in data compared to previous years. Some few municipalities might report data with errors where it is difficult to detect the errors. This is particularly true if data are stable over time or the level of the data seems reasonable in comparison with other municipalities. Some municipalities change their way of reporting data to Statistics Denmark or realize that previous data have not been reported correctly. Further more change of systems in the municipalities might cause changes in the reported data. This might cause fluctuations in data over time for certain municipalities. It is not possible to carry out further checks on the reported data otherwise approved by the individual municipality.

The quality of data that are collected by questionnaire 78 is subject to greater unreliability due to that is it difficult to collect the information from the municipalities. This is related to data on foster families that are published in RESFAMPL in the Statbank, private foster care and residential institutions for children and juveniles (RESFDPJ), home training of children and juveniles (RESHJMTR), attendant scheme for children and juveniles in special need for assistance (RESLED) and special institutions for children and juveniles (RESSBU2).

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

Revisions of data from previous years will be updated in Statbank Denmark when data are updated once a year. How many years data are corrected back in time depends on the possibilities of getting updated data back in time.