Statistical processing

Source data

The main source for compiling the statistics is accounts data for farms, whose accounts are prepared by accounts offices, which are organized within SEGES innovation. The data for the statistics are collected electronically from the accounts system Ø90, where additional and necessary information is supplemented by the accounts offices. In addition to this, accounts data for a number of enterprises, mainly gardening are collected from private accounts offices, which report the data electronically on questionnaires. Another source is the Farm Structure Survey maintained by Statistics Denmark. The register forms the basis for determining the population. Finally, register-based data on paid-out subsidies from the Danish Agricultural Agency are used.

The selection plan is optimized according to the Neumann-criteria, including three target variables. The Neumann allocation is based on the observed spread within strata defined by EU-type * size. The three target variables are Net profit, Debt ratio and investments in agricultural assets and weighted, respectively ½, ¼ and ¼. The selection is divided into 6 main types of agriculture; conventional full-time and part-time holdings, ecological full-time and part-time holdings, and full-time and part-time horticulture.

The Danish sample to FADN is in regulated to 1.600 farms.

Frequency of data collection

Data are collected yearly.

Data collection

Accounts offices, which are organized within SEGES Innovation, report data electronically from the accounts system Ø90, where additional and necessary information is supplemented by the accounts offices. In addition to this, accounts data for a number of enterprises, mainly gardening are collected from private accounts offices, which report the data electronically on questionnaires Reporting form. The information from Ø90 is collected as system-to-system reports. Finally, register based information is collected from the Register of Business Statistics by Statistics Denmark, Kopenhagen Fur and subsidies from the Danish Agricultural Agency.

Data validation

Before accounts are delivered to Statistics Denmark there are a number of tests, also stop-tests, in the accounting system Ø90 and in the questionnaires. After receiving the data at Statistics Denmark a sophisticated test is performed for each holding. The test is particular sophisticated because individual data are delivered to FADN, DG Agri.

Data compilation

The sample is weighted by use of the SAS program CLAN.

The classification according to the economic size is made on the basis of the total standard output (SO) of each individual farm. The SO is calculated for all types of crops and species of livestock as an average of the actual outputs (correspond to the Gross Output) per hectare or per animal in a reference period of 5 years. In former publications the standard gross margin (SGM) has been used. The difference between SO and SGM is that SGM reflects the difference between the gross output and the associated costs per units whereas SO only contain the gross output. In most cases, SO should be larger than SGM however the size of the subsidies which some activities contain can cause SGM to be larger than SO. SO is composed of the different activities on the farm and the type of holding is determined. As a main rule a type of holding is determined when 50 pct. of the total SO is from one specific activity.

In selecting the farms, we have made efforts to include the largest possible number of farms several years in succession. In practice, this implies that a maximum of 80 pct. of the agricultural farms that were included in the statistics for a given year were also included in the previous year. For different types of gardening it is not possible to limit the number of repeats because the number of farms in the sample will be inadequate.

Compilation of the accounts statistics

The average figures per farm for the total population and for sub-groups are estimated as a weighted average. For all agricultural farms and most groupings, a classification into sub-groups according to economic size is used. In some tables, a quartile distribution for central figures (e.g. the net profit) has been conducted for the purpose of analysing the variation of the key figure in question. The information stated is an average of, respectively, the 1st and 4th quartile group - and not the result of the farm, which is situated on the quartile limit.

Adjustment

None.