Skip to content

Accuracy and reliability

Contact info

Food Industries, Business Statistics
Henrik Bolding Pedersen
+45 39 17 33 15

hpe@dst.dk

Get as PDF

Accounts Statistics for Agriculture

The statistics are compiled on a sample population and consequently, the results are subject to some degree of uncertainty. The sample is stratified with the aim of being representative for all farms. From the population of small farms, where the variation is small, a relatively small sample is selected, while from the population of large farms a larger sample is used. Loss of sample farms are countered by higher selection rates in strata where losses do occur (based on experience).

Overall accuracy

The statistics are compiled on the basis of a sample population and consequently, the results are subject to some degree of statistical uncertainty, although the data extract is representative with a stratification taking into account that all farms are represented. The statistical uncertainty differs for each individual item, and the largest degree of uncertainty is seen for the item investments, which may vary considerably among the farms over time. Participation in these statistics are voluntary for the farmer which adds to the uncertainty.

Sampling error

Overall accuracy: The statistics are compiled on the basis of a sample population and consequently, the results are subject to some degree of statistical uncertainty, although the data extract is representative with a stratification taking into account that all farms are represented. The statistical uncertainty differs for each individual item, and the largest degree of uncertainty is seen for the item investments, which may vary considerably among the farms over time.

The overall accuracy is considered high. From the population of small farms, where the variation is small, a relatively small sample (1 to 2 pct.) is selected, while from the population of large farms, where the variation is greater, a sample of up to 20 pct. is used. In the case of special types of farming, e.g. poultry additional agricultural holdings are selected in order to be able to show reliable results.

Non-sampling error

In relation to the target population, which is Danish agriculture, more than 99 per cent of the Standard Output is covered. Selection in each strata is increased to facilitate loss of responses. In 2020 loss of responses was 9-13 per cent. Some of the loss is due to that the population is not known at the time of selection. Farm can fx be sold or go bankrupt in the time from selection to the actual accounting 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

Coverage and sample survey

The sample survey covers about 9 pct. of the entire population within agriculture as a whole. If the sub-populations are considered, the selection ratio, however, varies considerably, partly to take account of the greater spread of results among the large farms, partly to achieve a sufficiently large number of farms in order to be able to represent the small sub-populations. In connection with the selection procedure, we have aimed at including the greatest possible number of farms in the statistics over several years in succession.

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

Preliminary data for agriculture are published in the beginning of July when 55-60 per cent of the data has been a first examination.