The first ever Agriculture Census of Tonga was conducted in 1985, followed by the second one in 2001, while this is the third.
Agriculture is the predominant activity in the Kingdom of Tonga's economy, contributing more than 17% to the Gross Domestic Product (GDP) in 2012 - 2013. The first ever Agriculture Census of the Kingdom was conducted in 1985. The second Census was conducted in 2001, focusing on land tenure, land utilization, area and production of principal crops, livestock, agricultural implements and equipment, use of fertilizers, etc. including the various agricultural activities in which most of the households were engaged in. Although agriculture is the main factor in the economy of the Kingdom of Tonga, the database in this sector seems to be inadequate. There were quite several surveys conducted for this sector, however, an updated frame (list of holdings/parcels and its characteristics) is needed so these surveys will obtain more reliable estimates. There were important developments in agriculture within the fourteen-year period from the last census that should be captured like the use of forest trees within the farming system to enhance productivity and information on fisheries, which is becoming a very important sector of the Kingdom's economy. Considering the above issues, there is a great need to update the statistics on agriculture in order to determine its present situation and to use it for economic planning and policy-making.
In support of the strategic plans and programmes of the Kingdom of Tonga on agriculture, the Government has decided to conduct the Agriculture Census (AC). This census is envisioned to:
a) Provide benchmark or basic data on structure of agricultural holdings and their main characteristics;
b) Use this information to develop a regular system of agricultural statistics;
c) Build up some important village level statistics;
d) Establish a technical and organizational foundation on which to build up a comprehensive and integrated system of food and agricultural statistics; and
e) Provide a frame from which samples can be drawn to study certain aspects of agricultural activities in greater depth.
f) Provide information on community (village) statistics.
Kind of Data
Census/enumeration data [cen]
Unit of Analysis
The 2015 census covers all households and agricultural holdings as well as all villages in Tonga. Items covered include the
1. AT THE HOUSEHOLD AND ORGANIZATIONAL LEVEL:
a) Engagement of subsectors of agriculture
b) Household membership demographic and economic information
c) Land use under bush allotment and town allotment, area by tenure and location
d) Area under perennial crops, annual crops, animal yard, pasture, fallow and other land use
e) Number of trees in town allotment and home garden
f) Food crops cultivation, including crop still growing, crops harvested
g) Agricultural practices
h) Engagement in farming community
i) Livestock including number of heads, livestock disposal and veterinary services
j) Fishery including type of fishing, main purpose of fishing, species harvested and habitat type of fishing
k) Forestry including intended use of trees and shrubs
l) Handicraft including proportion of raw material used from household and estimate of product value of katoanga
m) Agricultural labour and machinery
2. AT THE COMMUNITY (VILLAGE) LEVEL:
a) Village officer's information
b) Main economic activity
c) Outbreaks in the village
d) Farmers' cooperative
e) Development priorities
Agriculture & Rural Development
Forests & Forestry
Land (policy, resource management)
Producers and sponsors
Tonga Statistics Department
Government of Tonga
Tonga Ministry of Agriculture, Food, Forestry and Fisheries
Government of Tonga
Statistics for Development Division
Technical assistance in data processing
Food and Agricultural Organisation
Tonga Ministry of Agriculture, Food, Forestry and Fisheries
Funding and Field operations
Tonga Statistics Department
Funding, Field operation and data processing
TECHNICAL NOTES AND DEFINITIONS
1. Agriculture active household is a household that is active in any of the agriculture activities: cropping; livestock;
fisheries; forestry and handicraft. A household is active in any of these agriculture activities if it can be classified into either: subsistence, semi subsistence or commercial.
2. A household is non active in agriculture if it cannot be classified into any of the agriculture activities: cropping; livestock; fishery; forestry or handicraft.
3. Subsistence is a type of agriculture activities (cropping; livestock; fishing; forestry or handicraft making) in which most of the produce is consumed by the farmer and his family, leaving nothing to be marketed.
4. Semi-subsistence is a type of agriculture activities in which some of the produces are to be consumed by the farmer and his family and some of them are to be marketed.
5. Commercial is a type of agriculture activities in which most of the produces are to be marketed.
6. For the 2015 TNAC, the following definition was used to classify the levels of agriculture activities whether it was subsistence; semi-subsistence or commercial.
Crop was based on total cultivated land area
a) Subsistence: 0 < and <= 1 acres
Semi Subsistence: 1 acres < and <= 8 acres
Commercial: > 8 acres
Livestock was based on type and number of livestock kept
a) Subsistence: If Milk Cattle or Beef Cattle =1 or Sow = 1
b) Semi Subsistence: If Milk Cattle or Beef Cattle = 2 - 100 or Sow = 2 - 25
c) Commercial: If Milk Cattle or Beef Cattle > 100 or Sow > 25 or Egg layer > 0 or Broiler > 0
Fishery was based households or organizations' response to the question on purpose of their fishing activities; whether it was for subsistence; semi-subsistence or commercial.
Forestry was based on number of high value trees and timber trees that households have grown at the time of the census
a) Subsistence: number of trees 1 - 4
b) Semi Subsistence: number of trees 5 - 100
c) Commercial: number of trees 101 - 999
Handicraft was based on proportion of handicraft being sold
a) Subsistence: no handicraft sold
b) Semi Subsistence: 1% - 75 % sold
c) Commercial: 76% - 100% sold
Deviations from the Sample Design
The final national response rate was 89% with 16,122 households enumerated out of the 18,043 total households.
Dates of Data Collection
Data Collection Mode
VERIFICATION AND CODING
The data editing process begins when the completed questionnaires were returned to the national statistics office (NSO) for checking by the coders. This include checking that all fields are correctly filled, skipped pattern are properly followed, missing fields and so forth. Once the questionnaires are verified to be correct, then coding begins where certain variables are coded to their respective codes for capturing in the data entry screen - codes include Village code, Crops and Tress codes.
The data entry application for the 2015 Agriculture Census was designed using the software package CSPro where all necessary checks were incorporated to allow the Data Entry Operators (DEO) to verify data while doing data entry. With all the in-built checks, this ensures capturing good quality data efficiently and effectively. The in-built checks include range checks, skip and filtering questions and consistent and logic checks. With these in-built checks ensures good quality data is captured while entering and this greatly helps in the final batch editing.
After the completion of the data entry, the final editing process was done. This include verification of questionnaires that all are captured and the actual running of the batch editing program on the whole data.
Since most of the checks were done during the data entry phase, the batch editing process mostly involves verifying
those errors that were missed or could not be solved during data entry, checking on those responses which have been
coded 'missing' and trying to impute or verify by referring to the respective questionnaire and fixing 'outliers' responses.
Frequencies on each variable were also checked to verify any inconsistencies between variables.
The batch editing logic program was ren twice when it was decided to finalize the data. Some missing values were not fixed as they were not able to be verified, examples of these are mostly on money values and number of crops/trees.
A Pilot Census was conducted which the questionnaires received were used to test the data entry application. This allowed to redefine the questionnaires as well as the data entry application to ensure that it everything was efficiently designed to capture reliable data. Like any other census, the 2015 Agriculture Census (AGC) has its own limitations. These are summarized as follows:
- The 2015 AGC only covered 89% of the total households in Tonga; 11% of the households were not enumerated.
- The 2015 AGC took place at the aftermath of Cyclone Ian which had destroyed a large part of the agriculture sector and the infrastructure for the Ha'apai Region.
- The 2015 AGC Questionnaire design meant that some of the data was not comparable to 2001. For example, the levels of agriculture activities (subsistence, semi-subsistence and commercial) and the way they were described.
- The 2015 AGC took place between April and June 2015 whereas the 2001 AGC took place between October and December 2001, therefore there is variation in the data due to seasonality of some of the agriculture activities, such as cropping and fishing.
- In the 2001 AGC the size of agricultural holding, to be considered as an agricultural holder was more than 1/8 of an acre. The 2015 AGC included town allotments below 1/8 of an acre. This was done so that agriculture activities in the town allotment could be accounted for.
- Questions on the use of fertilizers were answerable only by "Yes" or "No" and questions on agricultural chemicals asked only for the name of the chemical and the crops they were used on. Quantities of these agricultural inputs were not gathered.
- In the 2015 AGC, the question on the level of agriculture activities was only asked for the fisheries sub-sector and not for the other sub-sectors. Therefore, these were determined by the size of the land cultivated, the numbers of livestock kept, the number of forest trees grown, and the number of handicrafts made and sold.
- The time allowed for data processing, editing and verification was rather limited and did not allow sufficient time for thorough and in-depth checking of the data.
All information with regards to a respondent or individual are kept confidential under the Census and Statistical Act.
All information are coded in such a manner that no household or individual would be easily identified.
All users of the data will have to sign an agreement or undertaking that:
1. Refrain from making copies or reproduce the data he or she is granted with.
2. Refrain from attempting to identify any person or household from the data he or she is granted with.
3. Refrain from revealing the identification of any individual or household in the data he or she is granted with
The Data is made available under the following conditions:
1. The data will be specifically used for statistical and scientific research purpose ONLY
2. The data is not sold or re-distributed to any other individual, institution or organization
3. Adhering to the confidentiality clause in the Act as well as the Agreement.
4. The source has to be acknowledged in all modes of presentation, books, articles, papers, theses, reports or any other publication
5. A copy of all reports, publications and presentations are to be forwarded to the owner of the data.
"National Statistics Office of Tonga, Agriculture Census 2015 (AGC 2015), Version 01 of the Licensed datasets (April 2015), provided by the Pacific Microdata Library. http://pdl.spc.int/index.php/home"
Disclaimer and copyrights
The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses