The agricultural census is based on an annual postal survey of farms. In England and Wales, the sample size for the June survey changes each year depending on UK and EU requirements. In years such as 2010 and 2013 when the EU requires very detailed information on the structure of the UK agricultural industry the sample size is increased, up to 100% of commercial holdings, in other years it can be as low as 30% of commercial holdings.
The results for non-surveyed farms are interpolated from the most recent data for those farms using the trends on comparable farms. The data series extends back as far as 1866, although the detail is reduced for the historic datasets.
The data can be provided at holding level, although this is typically only for government / policy research and on the condition that the data, or data derived from it, is not presented in such a way that details are disclosive for a single or small number of holdings. Summarised data can be provided for specific areas, but constraints on disclosive data and use of the data would still apply.
Geographical breakdowns for England are readily available in the years that correspond to the EU Farm Structure Survey (most recently 2013 and 2016). Data for local authorities are summarised by land use and livestock (e.g. cattle), whilst data at county level is summarised by crop type and a slightly more detailed livestock type (e.g. dairy breeding herd, breeding pigs). Disclosive data items are not included.
Annual census of agricultural and horticultural activities for England, Scotland and Wales is conducted on commercial holdings, defined as holdings of more than 5 ha agricultural land, 1 ha orchards, 0.5 ha vegetables or 0.1 ha of protected crops and ornamentals, or more than 10 cows, 50 pigs, 20 sheep, 20 goats or 1000 poultry. Note that these thresholds have only been in place since 2010. Each farmer declares the agricultural activity on the land via a postal questionnaire. Since the late 1990s the agricultural census in England and Wales has been conducted as a sample survey in which data is only sought from a proportion of holdings in each year, as low as 30% of commercial holdings in some years (approx. 30,000 holdings). Since 2007, cattle numbers have been taken from the Cattle Tracing Scheme. Data for non-surveyed farms are interpolated from the most recent data for those farms using the trends on comparable farms.