Agricultural, Forestry and Environmental Statistics Department

Na padesátém 81

100 82 Praha 10

The Farm Structure Survey 2016 was preceded by the following surveys organized in line with the EU legislation: sample survey FSS 2013; full survey Agrocensus 2010 (= FSS & SAPM 2010); sample surveys FSS 2007, FSS 2005, and FSS 2003; and full survey Agrocensus 2000 (=FSS 2000). The first survey of this nature conducted in the Czech Republic was Agrocensus 1995. It was focused on description of deep structural changes in the Czech agriculture due to fundamental economical transformation which followed the change of the political orientation in 1989.

Former Czechoslovakia provided FAO with data from regular exhaustive surveys in the frame of the world censuses of agriculture in 1970, 1980, 1990; no special surveys were organized during this period. In 1930 a census of agricultural works was conducted as a part of world agricultural census coordinated by the International Institute of Agriculture in Rome.

^*) In the Czech Republic the term "agricultural censuses" is  traditionally used not only for exhaustive surveys but also for particular sample surveys including FSS. 

EC                   European Community  

EU                   European Union

EUROSTAT       Statistical Office of the European Communities

FAO                 Food and Agriculture Organisation of the United Nations

FSS                 Farm Structure Survey

LPIS                Land Parcel Information System

LSU                 Livestock unit

NACE               Statistical Classification of Economic Activities

NUTS               Nomenclature of Units for Territorial Statistics

RDM                Rural Development Measures

SAPM              Survey on Agricultural Production Methods

SO                  Standard Output

UAA                Utilised Agricultural Area

- growing of non-perennial crops (NACE Rev. 2 code 01.1);

- growing of perennial crops (01.2) including wine production from self-produced grapes;

- plant propagation (01.3);

- animal production (01.4); agricultural holdings raising other animals (01.49) are excluded except (i) ones raising and breeding ostriches, emus and rabbits, and (ii) ones dealing with bee-keeping and production of honey and beeswax;

- mixed farming (01.5);

- support activities to agriculture and post-harvest crop activities (01.6): only holdings exclusively maintaining agricultural land in good agricultural and environmental condition (01.61).

The Farm Register includes holdings reaching the following threshold levels:

- 1 ha of utilised agricultural land own or rented, or

- area of 1 500 m² of intensive crops (orchards, vegetables, flowers), or

- 1 000 m² of vineyards, or 300 m² of greenhouses or hotbeds, or

- 1 head of cattle, or 2 head of pigs, or 4 head of sheep and goats, or

- 50 head of poultry, or 100 head of rabbits or fur animals.

As at the day of generation the population in the Farm Register contained 47 416 agricultural holdings. 

- Area ≥ 5 ha of utilised agricultural area (own or rented) [A_3_1$ha];

- or ≥ 1 ha of orchards [B_4_1$ha];

- or ≥ 3,500 m² of vineyards [B_4_4$ha]

- or sum of area for vegetable, strawberries and flowers [B_1_7$ha+B_1_8$ha] ≥ 2 500 m²

- or ≥ 5 head of cattle [C_2$heads], or ≥ 10 head of pigs [C_4$heads],

- or ≥ 10 head of sheep and goats [C_3_1$heads+C_3_2$heads];

- or ≥ 100 head of poultry including ostriches [C_5$heads].

Geographical coverage: the whole Czech Republic

1)     12 sampled holdings (12 holdings in the population) have only animal production, especially other poultry.

2)     10 sampled holdings (67 holdings in the population) have only crop production. These holdings have fallow land and/or permanent grassland no longer used for production maintained in good agricultural and environmental conditions.

3)     1 sampled holding (12 holdings in the population) does not have any agricultural activities (not even kitchen garden). This holding grows short rotation coppices that are recorded as utilised agricultural area in the Czech Republic. Thus, it was included in the population by accident.

[1] Livestock unit means a standard measurement unit that allows the aggregation of the various categories of livestock in order to enable them to be compared. 1 Livestock unit represents 500 kg of live weight of an animal.

(ii) If the holding's headquarter ≠ the place of its agricultural production, then the location is the Territorial Technical Unit^[1] where the largest parcel is placed (2 205 holdings).

[1] A Territorial Technical Unit is a unit defined as cadastral territory or part of cadastral territory separated by the border of a basic settlement unit, i.e. by the border of a municipality, city district or city part. A municipality (LAU2) consists of one or more Territorial Technical Units.

Land use variables: 1 October 2015 – 30 September 2016

Livestock variables: 30 September 2016

Support for rural development variables: last three years (2014 – 2016)

1.    Preparation, organisation, coordination, monitoring

2.    Field survey

3.    Processing of the data

4.    Data validation and dissemination

As similar surveys were conducted in 2003, 2005, 2007, 2010 and 2013, there was no pilot survey organized in the frame of FSS 2016.

Timetable:

o   September 2015:

• Signature of the project agreement

o   October 2015 – May 2016:

• Testing of questionnaires
• Drafting of cover letters for municipal offices and respondents.
• Drafting of contracts for work and contracts for services.
• Printing of questionnaires with methodological notes
• Development of population and sample, application of precision requirements

o   June 2016:

• Staff preparation meetings
• Training of staff of Field Surveys Units and
• Working out of Minutes of the staff training meeting including explanatory notes and examples
• Programming of enter data, checking, control tables

o   July – August 2016:

• Training of interviewers at regional offices
• Providing methodological service

o   September – November 2016:

• Field work, collecting questionnaires, data checks
• Providing methodological service

o   December 2016:

• Finalisation of field work
• Working out of a conversion between the questionnaire structure and Eurofarm structure
• Paying-out of the interviewers

o   January - February 2017:

• Finalisation of questionnaires’ recording mistake corrections, check runs
• Reinvestigation of missing data at regional offices

o   March – August 2017:

• Programming of outputs tables
• Compiling a programme for conversion of data in Eurofarm format
• Data validation and verification

o   24 August 2017: Data dissemination (press conference, electronic and printed publications)

o   September 2017:

• Asking for administrative data
• Calculation of SO 2013 coefficients
• Working out of National Methodological Report

o   October 2017:

• Transmission of microdata to Eurostat
• Calculation of SO 2013 coefficients
• Working out of National Methodological Report

o   November 2017:

• Calculation of SO 2013 coefficients
• Working out of National Methodological Report

o   December 2017

• Transmission of NMR to Eurostat

o   Turn of Q1/Q2 2018

• Final technical implementation report and financial statement

1) Agricultural, Forestry and Environmental Statistics Department

- project administration unit

- responsible for the organisation, coordination, financial and methodological support

- generation of population and sample

- grossing up procedures

- data checks based on Eurostat validation rules

- processing of output tables

- dissemination of survey results

- delivery of the database of anonymized individual data to Eurostat.

2) Field Surveys Units

- recruitment and training of the interviewers,

- distribution of questionnaires to the interviewers and their methodological support,

- collecting of filled-in questionnaires, their coding and registering,

- data processing (data entry and checking using Dante software)

3) Unit for Processing of Agricultural and Forestry Indicators in Kind

- questionnaire collection from legal persons and large natural persons,

- centralised processing of questionnaires

- processing of questionnaires

- checks for correctness and completeness of the data files at the national level

- development of net database and its delivery to the Agricultural Statistics Dept.

Institute of Agricultural Economics and Information:  SO 2013 computation, classification of holdings by their type of farming and economic size

Variables on Rural Development Measures were taken from an administrative source (see item 4. below)

It is a list frame.

The Farm Register is updated regularly. Data on livestock numbers and areas under crops are updated yearly after finishing particular statistical surveys. Data in the Farm Register are not subject to any quality control.

The Farm Register was updated on the basis of FSS 2016 results as well as results of other surveys.However, the Farm Register covers agricultural holdings where the agricultural area utilised for farming is one hectare or more (see Article 3, par. 1a), 1b) of Regulation (EC) No. 1166/2008 of European Parliament and of the Council). The average area per holding is 73.4 ha.

In addition, administrative sources are used for Farm Register update: (i) Land Parcel Information System (LPIS) and (ii) Livestock Register. The tables below compare data from the Farm Register and administrative sources.

Comparison of sources – utilised agricultural area (ha)

Sources:

1) Czech Statistical Office (CZSO); updated on the basis of FSS 2016 results and Final Harvest Figures

2) Ministry of Agriculture, data as at 11.1.2017

3) Czech Office for Surveying, Mapping and Cadastre (COSMC) - administrative data source; the land is registered as agricultural land with its sub-categories irrespective whether it is utilised or not; data as at 31.12. 2016

Comparison of sources – livestock (head)

Sources:

1) Czech Statistical Office (CZSO); updated on the basis of FSS 2016 and Livestock Survey results

2) Ministry of Agriculture, data as at 31.12.2016

The sample was defined as stratified random sample from the population based on affiliation of each unit to defined grossing-up group (stratum). Specific sampling rates for all units in a group (stratum) were assigned to each stratum. Units with sampling rate under 100% were selected from the population using random sampling without replication. The units selected were not put back to the population, i.e. every unit could have been chosen only once. In consequence, the probability of selection of units in the population changed after each sampling.

2. Agricultural production areas which are defined on the basis of their agro-ecological and economical conditions.
In the Czech Republic are five areas: maize area, sugar beet area, potato area, potato-oats area, and mountainous area.

-      Maize area and its subareas: codes 11, 12, 13,

-      Sugar beet area and its subareas: codes 21, 22, 23,

-      Potato area and its subareas: codes 31, 32,

-      Potato-oats area and its subarea: code 33,

-      Mountainous area and its subareas: codes 41, 42.

Due to insufficient total area in subareas 11, 12, and 13, units belonging to subareas 12 and 13 were affiliated to subarea 11. Similarly the subarea code 42 was replaced by 41. Therefore only codes 11, 21, 22, 23, 31, 32, 33 and 41 were allowed.

3. Size of the holding 

On the basis of above-mentioned criteria every unit from the population was assigned with a six-digit code XXYYZZ which defined its affiliation to particular stratum:

XX = size category (11, 12, 13, 14, 15, 99);

YY = abbreviated code of NUTS3 (11, 21, 31, 32, 41, 42, 51, 52, 53, 61, 62, 71, 72, 81);

ZZ = agricultural production area (11, 21, 22, 23, 31, 32, 33, 41).

Number of grossing-up strata constructed using the described way resulted in 524 combinations, although the theoretical maximum of combinations was 672 (6 x 14 x 8). The final database sent to Eurostat includes 524 strata.

Variable used for calculation of the standard deviation within strata (for the optimal allocation): size of the holding (UAA, LSU); different costs across strata were not considered.

The resulting sample size was based on threshold levels for the 100% strata and on percentages of particular size categories. They were set on the basis of the methodology used for FSS 2010 sample. The final values were adjusted in order to reach the best homogeneity of the stratification categories.

The construction of the strata was analogous to that used in FSS 2013. The percentages of size categories were set in the same way as in other sample surveys.

Description: The State Agricultural Intervention Fund is an accredited paying agency - an administrator of financial subsidies both from the European Union and from the national financial funds.  Within the scope of the Common Agricultural Policy, the financial subsidy is granted from the European Agricultural Guarantee Fund (EAGF), from the European Agricultural Fund for Rural Development (EAFRD), and, within the scope of the Common Fisheries Policy, the European Maritime and Fisheries Fund (EMFF). The new Programme for Rural Development (RDP) drawing out financial resources from the EAFRD has replaced the Horizontal Rural Development Plan (HRDP) and the Rural Development and Multifunctional Agriculture Operational Programme (OP RDMA).

Legislation frame: The State Agricultural Intervention Fund is a legal entity with its residence in Prague and its activities are regulated by the Law No. 256/2000 Coll., concerning the State Agricultural Intervention Fund, as amended by the Law No. 128/2003 Coll. and the Law No. 85/2004 Coll. and the implementing regulations in the form of the Czech Government Order.

Time reference: 2014-2016

Updating of the source: after each approval of particular request.

The CZSO does not participate to the conceptual design and subsequent related revisions of the administrative sources.

Common identifiers: Unique identification number of a holding (a person in case of unregistered natural persons).

Method(s) of integration: linkage of the FSS and RDM databases using the unique holdings’ ID. In several cases the linkage for particular holding had to be searched in both databases step-by-step.

Main problems were connected with creating simple links between databases caused by lack of common identifiers.  In several cases it was not possible to create the linkage for particular holding automatically and both databases had to be searched step-by-step. 

1) Questionnaires sent to respondents using data boxes^*) (2 856 holdings of legal persons and 1 056 holdings of natural persons);

2) Questionnaires sent by regular mail (46 holdings of legal persons and 4 554 holdings of natural persons);

3) Face-to-face interviews (3 449 holdings of natural persons).

Online questionnaires were available for all holdings.

^*) A secured IT tool used for receiving and sending official documents in electronic form; according to the Czech legislation, its use is obligatory for legal persons but it is voluntarily used also by natural persons.

• Electronic data capture from filled-in paper questionnaire;

• Entering the data online by the holder.

Although the Law 89/1995 Coll., on the State Statistical Service, sets down rules for penalties to be applied in case of not providing requested data, the CZSO has never applied this approach. 

In case of a census 1=3+4+5

Only for sample survey, in which case 2=3+4+5

This item is a subset of 3.

4>4.1+4.2

5=5.1+5.2

5.1>=5.1.1+5.1.2

6=5.2+5.1.2+4.2

completeness checks,

relational checks,

data format checks,

range/outlier checks

Tools used for data validation

Data Supplier Manual validation rules developed by Eurostat.

Level of data validation

1)  basic checks by interviewers during the collection of questionnaires in the field

2)  checks at questionnaire delivery by external interviewers at the CZSO regional offices

3)  checks during the processing (logic checks, checks for completeness, linkage to nomenclatures, duplicities, etc.),

4)  validation controls by the Dept. of Agricultural Statistics. 

No characteristics not collected for other reasons occurred in FSS 2016. 

No cases where precision requirements are applicable and estimated RSEs are above the threshold occurred in FSS 2016.

In the grossing-up procedure weight=1 was assigned to each of these holdings. 

During calculation of relative standard errors 30 units showing extreme values were identified; their weight was changed to 1 and the units were moved to 100% stratum.

These holdings were included in the population on the basis of their last available address. The CZSO staff made attempts to find new adresses using e.g. the Business Register, LPIS, information from local authorities etc. but not always succesfully. Such holdings were classified as ineligible.

The data for non-responding holdings were imputed using two ways:

(i) Holdings with high similarity in the frame of their strata: average values for each stratum;

(ii) Holdings with extreme values: the last data available (i.e. from FSS 2013 or FSS 2010, LPIS, Livestock Register).

No comparison with another source or consistency study was conducted. Since the non-response rate is very small (1.58%) the non-response bias is expected to be negligible.

In case of missing data, the responding unit was contacted again and the data was filled in.

Corrections were made also by the staff of Field Surveys Units.

Tools used: Dante software, Microsoft Visual FoxPro

Day of the first publication of results: 24 August 2017

Time lag = 11 months (328 days) from the last day of the reference period.

Time lag related to the last day of the survey year (31 December 2016) = 8 months (236 days)

Time lag related to the last day of the survey year (31 December 2016) = 8 months (236 days)

The threshold was set up at such level, that only small agricultural holdings representing up to 2% of utilised agricultural area and up to 2% of livestock units[1] were excluded (see Article 3.2 of the Regulation No 1166/2008).

[1] Livestock unit means a standard measurement unit that allows the aggregation of the various categories of livestock in order to enable them to be compared. 1 Livestock unit represents 500 kg of live weight of an animal. 

• EU definition: hops [B_1_6_2] are included in crops grown on arable land;
• National definition: hop-gardens are reported as permanent crops according to the Law No 252/1997 Coll., on agriculture, as amended.

Data on total arable area and permanent crops in Eurostat database differ from the data published on the national level. Data on total utilised agricultural area (or permanent crops plus arable land) are the same.

• EU definition: nurseries [B_4_5] are included in permanent crops;
• National definition: nurseries are reported as crops grown on arable land.

Data in Eurostat database differ from the data published on the national level likewise.

• EU definition: trees and bushes grown for use for energy production only [B_6_3] are included in wooded area (of which short rotation coppices);
• National definition: they are included in other permanent crops (of which short rotation coppices).

Therefore in the Eurofarm database total utilised agricultural area is lower and wooded area is higher than in national data.

• EU definition: in the frame of organic farming, temporary grasslands are included in permanent grasslands;
• National definition: they are included in plants harvested green on arable land (perennial green fodder).

Number of hours per year for a full-time employee for AWU calculation: 1 800

• In 2016, 2013 and 2010 higher threshold levels were applied than in 2007, 2005 and 2003.
• In 2016 and 2013 different methods of sample construction were used than in 2007, 2005 and 2003.
• No changes in the coverage were introduced between 2013 and 2016 as well as between 2010 and 2013.

• A_3_2_1 The total utilised agricultural area of the holding on which organic farming production methods are applied and certified according to national or European Community rules
• A_3_2_2 The total utilised agricultural area of the holding that is under conversion to organic farming production methods to be certified according to national or European Community rules
• FSS & SAPM 2010: including kitchen gardens.
• FSS 2016 and 2013: kitchen gardens were not included (Handbook CPSA/SB/652, rev. 10).

There have been sufficient changes to warrant the designation of a break in series:

• C_5_2 Laying hens and C_5_3 Other poultry
• FSS & SAPM 2010: Pullets were included in Laying hens (Handbook CPSA/SB/652, rev. 6: „Growing pullets before point of lay are included if they have been stalled up as laying hens. Cull hens are also included. All hens which have started to lay are included, whether the eggs are for consumption or for breeding. Breeding cocks for laying hens are included“.)
• FSS 2013: Pullets were included in Other poultry (Handbook CPSA/SB/652, rev. 10: “Poultry not mentioned under Broilers or Laying hens. Chicks are excluded“).
• FSS 2016: A new characteristics C_5_3_x Pullets was introduced by Eurostat and used by CZ.

FSS 2016 and FSS 2013: sample survey

Changes possibly attributed to sampling variability: The sample was constructed using UAA and LSU (or particular livestock numbers) as stratification variables. During the grossing-up procedure emphasis was put on these variables; it means that for other variables the data may minutely differ from those that would be obtained in case of census.

Comparison of FSS 2016 results with data from the Livestock Survey as at 1 April 2016:

 FSS 2016 data are not reconcilable with data from other domains.

The publications are usually available in both paper version and electronic form at the website of the Czech Statistical Office (www.czso.cz). All publications are bilingual (Czech and English) and they are gratis (paid from a grant budget).

Farm structure survey 2016: output tables for the whole agricultural sector and for selected types of legal forms, basic information on the survey, methodological notes.

Farm structure survey 2016 in regions: basic information for regions (NUTS2) for the whole agricultural sector and by two main legal forms, methodological notes (only electronic version available).

Farm structure survey 2016 – analysis: farm typology, regional results, comparison of results of FSS 2016 and FSS 2000 (co-author: Institute of Agricultural Economics and Information). 

Farm structure survey 2016: 24 August 2017

Farm structure survey 2016 in regions: 15 December 2017

Farm structure survey 2016 – analytical evaluation: 27 April 2018

Farm structure survey 2016 in regions: https://www.czso.cz/csu/czso/farm-structure-survey-regions-2016

Farm structure survey 2016 – analytical evaluation: https://www.czso.cz/csu/czso/farm-structure-survey-analytical-evaluation-2016

Data confidentiality during dissemination is ensured by internal directive of the CZSO which states that statistical information can be released in case it is produced as an aggregation of at least three subjects. Furthermore, sum of shares for the two greatest individual values cannot exceed 85 % of the total aggregation. In case of exceeding this value the data are treated as confidential. Aggregation on a superior level must be done or another solution must be applied. The data can be released or provided to a third party on a contractual basis if the dominant subject provides a written approval to the CZSO.