Statistical processing

Source data

The information comes from the questionnaire, IACS and the livestock register In 2023 the sample had a size of 8.500 farms, about 40 per cent all farms in Denmark.

The sample is stratified by regions, size and type of farming (e.g. pig farms, cattle farms and plant production).

Frequency of data collection

The survey is yearly.

Data collection

The farmer reports the information on a web questionnaire on Virk.dk

Information on the survey can be found here: Information

The text is Danish.

Data validation

The statistics go through a validation process which falls into three steps:

1. In connection with the online reporting it is secured that all questions are answered. It could for example be yes/no questions where a farmer e.g. might answer that he does not have any livestock on his farm.
2. In the next step all the answers are treated and are subject to a plausibility check. Errors are corrected immediately, sometimes after telephone contact with the farmers.
3. In the last step the results are analysed by regions and extreme developments – big increases or decreases – are identified. In this step it is often possible to find a few but rather big mistakes which for one or any reasons have been overlooked in the first two steps.

Data compilation

The extrapolation into a total level takes place by obtaining certain known targets:

• Number of farms and area are distributed by size based on the farmer’s application for agricultural subsidies. (The single payment scheme).
• Number of pigs in pig surveys April and July same years.

In case of non response the farm is excluded from the sample and the extrapolation is increased accordingly for stratum where non response occurs. The non-response was about 1,0 per cent, in 2019.

Both in 2019, 2021 and 2022 most farms smaller than 25 hectares are imputed. These farms constitute about 50 percent of all farms.

2020 was special by being a total census where in principle all farms answer and non response does not exist. That is not possible in practice so about 9 percent of the farms were imputed in 2020. However,, they have only about 2. per cent of the agricultural area so it is mainly small farms which are imputed.

Adjustment

No corrections are made besides what is described in data validation and data compilation.