Accuracy and reliability

Overall accuracy

Thorough maintenance is carried out on an ongoing basis on the population used for enumeration, which is therefore judged to reflect reality well and the sample is taken in a way that ensures that there is a high coverage of all industries. Lapse, however, means that at the first publication there may be greater uncertainty regarding. precision, but is compensated for skewed drop-out in the enumeration, which includes information on the entire population from the quarterly statistics for Industry's sales of goods. When the month is revised for the second time, however, there is a coverage of almost 100%, which is why this uncertainty disappears. It can also be skewed that the very small companies are not selected for the sample. If those companies have a markedly different development than the remaining companies, then it can affect the precision.

Sampling error

The sampling uncertainty is due to the fact that the companies that must contribute to these statistics have been selected at random, and therefore could just as well have been some other companies. The sampling uncertainty can be quantified by the coefficient of variance, which is a relative measure that indicates how large a proportion of the estimate is made up of the standard deviation. The coefficients of variation for the total domestic market turnover is 1.1%, and for the total export turnover the coefficient of variation is 0.6%

Non-sampling error

There is no reason to believe that there are particular problems with errors in the reported turnover figures and stock figures. In the calculation of production, the change in inventory value is recognized in the turnover that is deflated. This method assumes that the price development in goods sold is the same as in stock goods, and this is subject to a degree of uncertainty. There may be different composition of the turnover in the individual companies and the price indices used for the quantity calculation, which means uncertainty in the calculated quantities. There are no companies with between 10 and 20 employees in the sample, and thus it is assumed that the development in these companies follows the development of the larger companies.

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

It is estimated that the statistics largely live up to the needs of users and that they live up to the EU regulation. The precision is judged to be high and the statistics are always published on the pre-announced date.

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

The monthly figures are revised twice, and revisions further back in time happen rarely (e.g. with the introduction of new methods). The difference between provisional figures and final figures is normally less than 1 percent. In some cases, however, revisions to information formerly reported by enterprises may lead to differences larger than that. The mean absolute revision (MAR) is calculated being 1,45 and the mean revision (MR) is calculated being 0,09 for the time series mining and quarrying, manufacturing and electricity, gas, steam and air conditioning supply (NACE BCD), seasonal adjusted. The size of the revisions is calculated on the basis of data for the previous base year (2015), as it requires a certain number of observations in the new base year (2021) before it is possible to calculate