In case of censuses an Act provides the legal background, while interim surveys are part of the annual data collection system which is a Government Decree in Hungary.

Agricultural enterprises are part of the business register of HCSO updated continuously with data transmitted from the Registry Court. The register of private holdings is updated fully on the basis of censuses and partly when interim surveys are carried out.

For the implementation of FSSs the combination of exhaustive and sample survey is used. Agricultural enterprises (units with legal entity) always observed on full scope while regarding private holdings (units without legal entity) based on their size surveys are carried out on full scope and representative basis.

For agricultural enterprises it is compulsory to report their agricultural activity via internet from 2013.

Compared to FSS 2013 two new elements were introduced regarding the data collection of private holdings in 2016:

1. in case of private holdings enumerators mainly carried out computer assisted face-to-face interviews (CAPI) by using tablet or laptop. (Part of the private holdings provided data electronically via internet with self-filling method as in 2013.)
2. key private holdings (big farms) had to provide their data electronically via internet with self-filling method.

The seven Regional Departments (NUTS 2 level) of the HCSO and County Sections (NUTS 3) belonging to them (these units constitute together the Data Collection Directorate – DCD as of February 2017) are involved in the implementation under the management of the Agriculture and Environmental Statistics Collection Department (AESD).

AESD – Agriculture and Environment Statistics Department

DCD – Data Collection Directorate

ADÉL – Uniform Data Entry and Validation System

CAPI – Computer Assisted Personal Interview

CVFP 2001 – Census of Vineyards and Fruit Plantations 2001

EAA – Economic Accounts for Agriculture

ELEKTRA – Internet based self-filling data collection application

E-MEZO – Internet based self-filling data collection application

EUF – European Farm Structure characteristic

FSS – Farm Structure Survey

GÉSA – Data Collection Organisation and Monitoring System of HCSO

HCSO – Hungarian Central Statistical Office

HPA – Hungarian Paying Agency

IACS – Integrated Administrative Control System

KARAT – Integrated data transmission system of HCSO

MA – Ministry of Agriculture (earlier: MARD – Ministry of Agriculture and Rural Development)

MD – Methodological Department

NFCSO – National Food Chain Safety Office

NPSDC – National Program of Statistical Data Collection

NCWC – National Council of Wine Communities

RD – Regional Department

SDC – Statistical Disclosure Control

Based on their legal entity the target population has two main groups in Hungary: private holdings and agricultural enterprises. 

• Private holdings: households engaged in agricultural activity reaching or exceeding certain physical thresholds at the reference time of the survey (see the thresholds below - item 2).
• Agricultural enterprises: legal entities engaged in any kind of agricultural activity or classified as agricultural producer by NACE regardless of its size.

The definition of agricultural holding in Hungary covers also

• holdings with only forest, fish ponds, reeds, fur animals other than rabbit;
• holdings providing agricultural services only;
• agricultural enterprises registered as agricultural producers in the Registry Court but without carrying out any agricultural activity in fact.

The holdings above the thresholds (see item 2. below) form the entire universe of the agricultural holdings in Hungary. The households engaged in agricultural activities but under the thresholds are not considered as agricultural holdings.

Of which: Key private holdings (reaching or exceeding certain physical threshold)

Agricultural enterprises:

Legal entities engaged in any kind of agricultural activity regardless of its size. Agricultural enterprises operated in 2016 formed the main frame of the survey based on the Business Register of HCSO. Additional agricultural enterprises were added based on administrative records.

• 1962 (sample) holdings with fallow land kept in good agricultural and environmental conditions;
• 133 (sample) holdings with permanent grassland and meadow - no used for production, eligible for subsidies;
• 52 (sample) holdings with ‘other livestock’ - rabbits except breeding rabbits;
• 506 (sample) holdings with area of kitchen garden and that of forest, fish-pond and reeds.

In case of common land units (5) the administrative centre of NUTS3 region is used.

1 June 2016 for land characteristics (including the changes of land area between 1 June 2015 and 31 May 2016),

1 June 2016 for livestock characteristics,

a period of 12 month ending 31 May 2016 for labour force and production methods characteristics,

Small private holdings: sample survey

Key private holdings: full scope survey

Administrative sources:

• Organic farming
• IACS
• Vineyard register

• Agricultural enterprises: list frame
• Key private holdings: list frame
• Small private holdings: combination of area and list frame

Agricultural enterprises:
The source of the frame is the Business Register of HCSO containing legal entities engaged in any kind of agricultural activity regardless of its size. Agricultural enterprises operating in 2016 constituted the frame. Additional enterprises were added based on administrative records (data of farmers receiving subsidy, organic farming register). Updated continuously with data transmitted from the Registry Court.

Private holdings:
The sources of the frame are:

• enumeration areas of AC 2010;
• Farm Register of private holdings;
  Updated exhaustively when an agricultural census is carried out (latest one was in 2010). After the agricultural census 2010 the Farm Register was updated only partially based on the information of FSS (2013) and regular sample surveys. In the preparation phase of FSS 2016 the register was updated with information from the following administrative sources:
• farmers receiving area based subsidies (from IACS);
• farmers involved in organic farming (from Organic Farming Register);
• licensed traditional small-scale producers (they are private persons involved in agricultural activities listed in the relevant law on his/her own farm and hold a registered licence for the activity and fall under special tax rules).

The next process was followed for the determination of the overall sample size. At first step the sample for private farms was considered as stratified sample, where the sampling units were survey districts and stratification meant geographical stratification. Summarized information about private farms was used at this step − in survey district level − based on the results of AC2010. All necessary input for the determination of the overall sample size (sum, average, variance) were calculated in this way.

In the second step was checked whether the sample is good enough as a cluster sample and satisfies the precision requirements. More than 100 sample selection and calculation of the RSEs for involved variables (based on the results of AC2010) were done considering our sample as a cluster sample.

Sample size is a number of survey districts (and not the number of holdings).
The sample contains 3 007 enumeration areas (districts).

Overall sample size were allocated amongs counties by a Neyman allocation, using the number of holdings for calculating the standard deviation.

• Legal base: Commission Regulation (EC) 889/2008;
• Time reference: 2016
• Updating of the source: continuously

Integrated Administrative and Control System (IACS)

• Legal base:  Council Regulation (EEC) 1782/2003               
• Time reference: 2016
• Updating of the source: continuously

Vineyard Register

• Legal base:  Council Regulation (EEC) 479/2008; Council Regulation (EEC) 436/2009
• Time reference: 2016
• Updating of the source: continuously

Uniform Animal Registration and Identification System

• Legal base: Ministerial Regulation (MARD) No 99/2002. (15. XI.) 
• Time reference: 2016
• Updating of the source: continuously

IACS and Organic Farming Register:

Beneficiaries in IACS, in Organic Register and FSS holdings are not identical.

IACS and organic production identifiers were collected on questionnaires. When several beneficiaries belong to one holding, aggregated administrative data were matched based on the collected identifiers. 

1. the land area of quality and other wines by wine communities have been received from the National Council of Wine Communities (NCWC), from which a ratio among them was calculated on the lowest available level,
2. the list of municipalities for each wine community is available,
3. the holdings cultivating vineyards were linked to the regarding wine community using the names (codes) of the settlements,
4. the vineyard area of each farmer in the regarding wine community was split among the quality and other wine categories according to the calculated ratio,
5. the vineyards outside the wine communities are considered as other wines, because legally quality wines are not to be produced on areas outside the wine communities.

The link between the organic farm register and the holdings surveyed is created by the following way:

• Matching statistical ID code in case of the agricultural enterprises;
• Matching IACS identifiers. Some respondents did not provide this data in FSS 2016 questionnaire;
• Matching organic producer ID code. Some respondents did not provide this data in FSS 2016 questionnaire;

Integrated Administrative and Control System (IACS) – agricultural holding

• Matching statistical ID code in case of the agricultural enterprises;
• Matching IACS identifiers. Some respondents did not provide this data in FSS 2016 questionnaire;

Vineyard Register – agricultural holding

Uniform Animal Registration and Identification System – agricultural holding

Matching name and address in case of the private holdings where IACS identifiers and organic producer ID codes were not available: in this case some manual work was needed.

• ELEKTRA (self-filling application)

From 2013 agricultural enterprises report their agricultural activity electronically within the ELEKTRA (https://elektra.ksh.hu/) system which is a legal requirement in Hungary.

Key private holdings were directed to the ELEKTRA system after closing the self-filling period. (They had the possibility to fill in their questionnaire during the self-filling period.)

• E-MEZO (self-filling application)

It is a web application accessible by internet which enables filling in the questionnaires electronically by data providers. The application includes controls within and between tables.

Certain part of private holdings (farms receiving subsidy, certified organic farms and licenced small-scale producers), together almost 51 thousand holdings and 3 thousand key private holdings could fill in the questionnaire electronically by using this method during the self-filling period. They were informed in a postal letter about their login code and password in the middle of May. By the end 15.1 per cent of these private holdings provided data with this mode.

• CAPI (Computer Assisted Personal Interwiev)

It is a web-based application which can be used both on line and off line modes and makes possible filling in the questionnaire by enumerators with mobile devices.

The questionnaires generated in the self-filling application and CAPI modules are stored in a central database. It contains the same controls as the E-MEZO self-filling application.

The communication campaign included the following elements:

• posters informing about the survey were placed in towns and villages;
• brochure by counties were produced including the most important information about FSS 2016, international and county specific data;
• press conferences: one held in the HCSO in Budapest and nine in different towns;
• announcements about the implementation of FSS 2016 were published in the nation-wide and local media, including official Facebook page of HCSO;
• articles and interviews relating to the implementation as well as the main features of the survey were published.

Call Centre: A Call Centre was operated between 10 May 2016 and 15 July 2016. Respondent could ask questions by using free telephone line. In total 3628 calls were received, 67 per cent during the internet self-filling period. Questions related to IT problems were answered by the staff of HCSO IT Services Department. 200 emails were received from the respondents in the self-filling period which were answered by the staff of AESD.

Private holdings

• The surveyors and supervisors were trained how to handle difficult respondents. The survey supervisors with the help of the local authorities managed to convince nearly all the non-respondents, thus legal steps were not taken.
• When holders could not be contacted, the enumerator left a note to inform the holder about the time of his/her next visit.

Agricultural enterprises

• Automatic reminders were sent via email to the enterprises (just before and after the deadline) that did not send the questionnaire back by due time.
• In ambiguous cases the missing data were fixed by the colleagues of the Regional Departments, by contacting the enterprises concerned.
• Legal steps were not taken.

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

FSS 2016 was carried out electronically in case of all holdings in Hungary. The advantage of the electronic data collection was that logical and arithmetical coherence checks within and between the tables were incorporated into E-MEZO (both the self-filling and the CAPI application) and ELEKTRA. They were the following:

• Completeness checks (e.g. registration of arrived questionnaire, causes of non-response);
• Completeness checks within the questionnaires (e.g. all identification information and compulsory information);
• Data format checks (ha, m²; use of decimals);
• Data value checks (e.g. number of working days cannot exceed 365);
• Logical and arithmetical coherence within and between the tables (e.g. coherence between arable land and irrigation);
• Routing (skip) checks (e.g. in case of no land use skip to the next question);
• Plausibility checks (e.g. current value must be within a plausible range from value in administrative data sources)  
• Questionnaire could be sent after all errors corrected.

Batch-checks:

Non-sampling error
During the survey and validation process the effects of non-sampling errors were reduced as much as possible. Therefore the importance of non-sampling error – apart from non-response error – was negligible. There was no significant step taken because of this reason. Non-response error was corrected by imputation.

Mandatory items only could be missed when the entire groups of questions were lost.

To prevent these cases admistrative data was used for checking inside the program in those cases where this was possible.

Private holdings: 2.4 %

All the data entry applications were developed by the IT Department in the framework of uniform Data Entry and Validation System run by the Central Statistical Office. As many validation rules as possible were incorporated to the data entry application such as logical and arithmetical coherence within and between tables, both in case of electronic and traditional data entry.

Four categories of error levels were handled during data entry phase as follows: (1) less serious errors only for information; (2) errors can be accepted, but justification is needed; (3) only authorized survey administrators can accept serious errors; (4) unacceptable errors must be corrected immediately, the data entry only can be carried on after correction.

Private holdings: 0,1 %

Final results: t+0

The number of hours/year for a full-time employee to calculate AWU: 8 hours/day; 225 days/year; 1800 hours/year.

The Hungarian legislation identifies precisely the land utilisation types and the users. Common land utilisation is not a legally existing form. Furthermore, for the eligibility for single area payment schemes aid, land use rights should be verified.

All of the local governments were involved (some 3 thousand data providers, but municipalities not carrying out any agricultural activity can have an exemption) in data collection for agricultural enterprises in the framework of the FSS2016 survey. Due to the fact, that the owner of a common land is a public entity, data providers could be limited to the group of agricultural enterprises and thus included in this survey.

During the survey, the following criteria were taken into account:

• only municipal grasslands are considered as common land;
• land is used by more than one holding simultaneously;
• common grassland is not divided, there is no territorial restriction for farmers;
• there can be an oral or written agreement, but it is not a must;
• utilisation is free of charge.

*The figures used in the table as common land refer to only permanent grassland and meadow - rough grazing.

The significant change of common land area from 2013 to 2016 can be explained by the followings:

• The measurement unit (hectare) was not understood by the respondents in 2013.
• The definition of common land was misunderstood by the holdings in 2013 as the results contained both owner and tenant farming compared to 2016 when the area used was registered.
• The methodology differs in some aspects, mainly in questioning. In 2016, common land was incorporated into the FSS questionnaire but only for grassland, in 2013 common land area was collected by all land use categories.

The FSS 2016 results have proved to be of good quality, however, the aggregates of different land areas cover only the area that can be connected to the agricultural holdings. (At the same time the current statistics cover the land area unidentifiable with holdings as well, which means that the published aggregates contain and reflect additional expert estimations.)

http://www.ksh.hu/docs/hun/xftp/idoszaki/gszo/agrariumelo16.pdf

Data compilation

http://www.ksh.hu/agrarcenzusok_agrarium_2016_tablak

Regional differences in the Hungarian Agriculture, 2016, final data

Anonymised microdata may be accessed by research institutions within the framework of a contract with the HCSO. Anonymised microdata are microdata which have been modified in order to reduce to an acceptable level, in accordance with current best practice, the disclosure risk of statistical units to which they relate.

The researcher may only use the data file for the scientific research purpose indicated in the request form and the data file has to be destroyed upon fulfilment of that purpose. The detailed conditions for the use of anonymised microdata are regulated in the contract between the research institution and the HCSO.

Researchers may also have access to microdata in a secure environment such as the HCSO’s Safe Centre, another remote access point and through remote execution. The Farm Structure Surveys data of 2000, 2003, 2005, 2007, 2010, 2013 are already accessible and that of 2016 are planned to be made accessible for researchers.

The HCSO facility in Szeged is providing a remote access service to researchers under the same conditions as for the Safe Centre environment (available on HCSO premises in Budapest).

In the form of remote execution, researchers can also apply for research outputs based on microdata sets. Using this access channel, the researchers are requested to send detailed specifications and descriptions, syntax, etc. to HCSO and the data is prepared by HCSO experts within HCSO. Outputs produced are released following an obligatory output checking procedure (common procedure for Safe Centre, remote access and remote execution).

Data made available in the Safe Centre, remote access and remote execution does not contain direct identifiers. The access environment is strictly monitored and the research outputs are checked for statistical disclosure before they may be taken from the safe environment by the researcher.

The process of the evaluation of the data request for the safe environment data access channels covers checking information both on the research purpose and the researchers. Access is granted based on a contract which stipulates the conditions of access.

• Act CLV of 2016 on Official statistics;
• Act CXII of 2011 on Informational Self-Determination and on Freedom of Information.

Besides the above mentioned legal acts, internal regulations on confidentiality exist within the HCSO. The access to statistical data is regulated in a separate internal regulation (Regulation 18/2014 on the rules of data access) which contains the rules on the six data access channels of the HCSO.

All research outputs produced in the safe environment (Safe Centre, remote access, remote execution) also go through obligatory output checking procedures.

Apart from these obligatory provisions, the typical SDC methods applied to FSS data are the following:

• global recoding (removing a dimension (e.g. column)),
• sub-sampling based on microdata,
• local suppression.