Census of Agriculture, 2010

(a) Frame
The FSS 2010 was intended for full coverage of the target population so that no sample needed to be drawn. This goal was achieved by dividing the total area of the country into two main parts:
urban and rural. In the case of rural areas, the survey was carried out from door to door, thereby guaranteeing an exhaustive coverage. All questionnaires completed in rural areas where also checked against the existing register mentioned above in order to identify and/or explain any changes in those holdings which were included on the register. Such checks were not possible, of course, in the case of new holdings for which no prior information was available. However, it must be pointed out that in the case of rural areas, all households were visited with the assistance of the community authorities and the local council. This ensured not only that all households were fully covered but also that the population in each village was encouraged by its local authorities to cooperate with the enumerators and to provide truthful information. In the case of urban areas this was not possible. The coverage in urban areas was based on the updated agricultural register. As mentioned above, the register of agricultural holders was created based on the census 2003, and was updated using FSS 2005 and 2007, other annual surveys and using information from a wide variety of sources.

(b) Survey design
The objective of the Census 2010 was to cover the population exhaustively. For this purpose the population was stratified by geographical area, that is, it was divided into two main strata: urban and rural areas. In the case of rural areas, exhaustiveness was achieved through the door to door process of data collection. In the case of urban areas, exhaustiveness was more difficult to achieve due to the fact that visits did not take place on a door to door basis. However, the large volume of information that was provided from the farm register assisted in minimizing the number of cases lost.

(c) Sampling strategy -for SAPM and/or OGA
As mentioned above, the sample was drawn using a stratified random sampling. According to the budget available to the Agricultural Sector of the Statistical Service of Cyprus, it was decided that the total sample size should be about 7.200 holdings from the total of 38.859 farms. More specifically the stratification and sampling procedure was based on the following steps:

1. At a first stage, a Standard Output for each holding was calculated using the standard output coefficients. These coefficients were applied on data collected by the Census on the utilized agricultural area of holdings and the number and kind of animals. These data were already available at the start of the SAPM survey since in May 2011, as the collection and data entry of the Census data had already been completed by April 2011.

2. At a second stage, all holdings were divided into 4 groups according to their Standard Output as follows: 0=SO=20.000, 20.000<SO=50.000, 50.000<SO=100.000 and SO>100.000. Then it was decided to take 1/7 from the first and second groups, 1/5 from the third group and 1/3 from the fourth group. At the end of the process, a sample was drawn comprising 7.146 holdings. Overall, the sample represented the 40% of total utilised agricultural area and the 47% of the total number of animals. More specifically, 46% of the 15 total area of cereals, 46% of the total area of potatoes, 52% of the total area of oilseed, 29% of the total area of permanent outdoor crops, 53% of the total area of vegetables, melons strawberries and flowers and 43% of the total area of temporary grass and permanent grassland were covered in the survey. Furthermore, 58% of total bovine population, 51% of total number of sheep and goats, 38% of total pig population and 47% of total poultry were also covered in the survey.