Accuracy and reliability

Overall accuracy

For employees in the private sector, the sample is drawn so that a large part of the target population is represented. For example are all enterprises with more than 100 employees (full-time equivalents) included in the sample. This means that the accuracy in industries with a high share of large enterprises have a lower uncertainty. For industries with a high share of small enterprises, the accuracy is lower.

Employees in enterprises with less than 10 employees are not covered by the statistics. Thus, employees in these enterprises can potentially have another development in average earnings than what is shown in the statistics.

To increase the accuracy, the standardised index of average earnings is delimited to exclude apprentices and young people under 18 years as well as cases where it can be difficult to calculate hourly earnings (e.g. persons paid in the form of fees or charges).

Sampling error

Sampling error has not been calculated.

Non-sampling error

There is some uncertainty in the information reported by private enterprises related to the completeness of data. Typically, it can be attributed to errors in the enterprises' pay systems, which ultimately generate data for the indices. For example, data might lack a division of earnings into the relevant individual parts, which causes an underreporting of the extent of irregular payments that are not included in the hourly earnings concept used. The extent is unknown.

There is some uncertainty regarding the correct industry and sector placement of some of the public employees, which arise due to linking the reported information with Statistics Denmark register for business statistics. The reason for this is that some employees in municipalities and regions are placed by industry in the municipal and regional houses, despite the fact that these personnel groups do not actually work in administrative functions.

A source of uncertainty in the private sector is that each quarter, approximately 10 percent on average of the sampled enterprises fail to report or are eliminated due to insufficient reporting. Missing data to this extent can affect the calculation of the index and cause bias. The magnitude of the bias due to non-response is unknown.

There can also exist some uncertainty related to employees in small enterprises with less than 10 employees. Enterprises with less than 10 employees are not reflected in the sample for the private sector due to the cost of reporting. Thus, employees in these enterprises can potentially have another development in average earnings than what is shown in the statistics.

In addition, it is not possible to adjust for the part of the development in earnings that is driven by increased skills or experience. Thus, there may be some uncertainty related to this.

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

Overall, the standardized index of average earnings is considered to be representative as an estimate for the development in average earnings for employees in Denmark. This assessment is based on the fact that the indices are based on an extensive data base, as it includes (more or less) all employment relationships in the public sector. For the private sector, the database consist of all employment relationships from a representative sample of enterprises where, in particular, all of the largest enterprises (with more than 100 employees) are included.

The different payroll elements that are reported to Statistics Denmark can be of varying quality. For example, the use of the DISCO-codes can differ among enterprises, which can affect the data quality. This can be difficult to adjust for, as the true DISCO-code is unknown. Typically, errors in DISCO-codes will not affect the standardised index of average earnings significantly, as errors are presumed to be within the same main DISCO function, which is the level at the index is disseminated at.

The reported earnings, on the other hand, will often be of high quality. This is due, among other things, to the fact that they undergo an indirect validation check, as the individual employee often will react themselves if the salary payment is incorrect. For the other payroll elements, such as the number of hours (and thereby the calculated hourly earnings), atypical reports will often be detected and subsequently handled as a part of the data validation process.

Moreover, missing reports from enterprises can also affect the quality of the standardised index of average earnings.

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

Only final figures are published. Revisions are only made when errors are detected.

In February 2022, the statistics were paused in order to conduct a comprehensive service review. As a result, the method for calculating the standardized index of average earnings was revised and the entire statistics have subsequently been recalculated. Figures published before February 2022 are thus no longer relevant.