Statistical Clearing House

Agricultural Survey, Financial Year 2006/2007

---

• Overview *** You are here ***
• Contact Information
  
• Summary
  
• Reference Period & Frequency
  
• Response Rates
  
• Clearance Information
• Quality Framework
• A. General Information
• B. User Requirements, Survey Objectives and Budget
• C. Alternative Sources of Data
• D. Data Outputs
• E. Questionnaire Design and Data Collection
• F. Respondent Load
• G. Population / Frame
• H. Sample Design
• I. Estimation and Analysis

Overview

---

Contact Information

	Portfolio:	Treasurer
	Agency:	ABS - Australian Bureau of Statistics
	Contact Person...	Name:	Margaret Dyas
		Email:	margaret.dyas@abs.gov.au
		Phone:	03 6222 5860 (International +61)

Return to top

Summary

Primary producers, industry organisations and suppliers, and other agricultural service and support industries use agricultural statistics for planning and decision making. The Federal and State Governments make extensive use of agricultural statistics for planning, budgets advisory services and for policies related to marketing agricultural commodities. The prime objective of the Agricultural Survey (AS) is to act as a source of basic agricultural statistics about a wide variety of commodities. The AS is designed primarily to collect data about commodities (including area and production for crops, number of livestock and area irrigated). Data are also available on the number of producers for each commodity, and on the counts of producers by Australian and New Zealand Standard Industrial Classification (ANZSIC) class. The AS is also the major source of commodity data input into the calculation of Value of Agricultural Commodities Produced (VACP). VACP is a major output from the survey which is in turn used as an input to the Australian National Accounts. The Agricultural Survey is the basic source of Australian agricultural commodity statistics. Frequency: The Agricultural Survey (AS) is conducted annually. Approximately every fifth year an Agricultural Census will be conducted in place of the AS. Scope: The frame population of the AS is all establishments with an Estimated Value of Agricultural Operations (EVAO) of $5,000 or more. The sample size for 2001/02 was approximately 35,000 respondents, and for 2002/03 it was 28,000. The sample for the 2003/04 Agricultural Survey is a combination of: (a) a reduced sample from the same selections as the 2001/02 Agricultural Survey (about 28,000 of the original 35,000 units); and (b) 3,000 new selections from the 2002 frame (to account for sample loss due to the increasing number of deaths). This extra sample will target industries where survey results for commodities are creeping out towards or beyond design parameters. Statistical unit: Data is collected at the establishment level. Information: The Agricultural Survey collects area and production data for a wide range of agricultural commodities. This commodity data is used to produce data on the Value of Agricultural Commodities Produced (VACP). Information on inputs to the production process is also collected and disseminated. Major Area of holding commodity Land use categories: Cereals and other broadacre crops Vegetables Fruit Vineyards Livestock Breakdowns: Data are available at State and Statistical Division (SD). Data are available by Estimated Value of Agricultural Operations (EVAO), Australian and New Zealand Standard Industrial Classification (ANZSIC), commodity and specified size ranges (e.g. herd size) and Area of Holding classes. Two complementary collections are run in conjunction with the Agricultural Survey, namely Apples and Pears (user-funded) and Vineyards (user-funded). The two user-funded surveys have their own separate entries. Design Constraints for major data items: Relative Standard Errors (RSEs) 2.5 % for Australian level 6% for State level 12% for regional level Minimum stratum sample size: 15 for size 1; 10 for others

Return to top

Reference Period & Frequency

	Reference Period:	Financial Year 2006/2007
	Frequency:	Annual

Return to top
Response Rates

			As planned	Actual
	Number of businesses...
	Contacted:		35000	Not available
	Live and in scope:		31500	Not available
	Responding:		25200	Not available
	Average time to complete questionnaire:		52 mins	Not available
	Statistics...
	Response rate:		80%	Not available

Return to top

Clearance Information

	Survey reviewed by
	Stat. Clearing House:	Financial Year 2006/2007
	Status:	Approved
	Approval period:	From 2006/2007 to 2008/2009
	For additional clearance information, please contact the Statistical Clearing House by phone, fax, post or email.

Return to top

---

Agricultural Survey, Financial Year 2006/2007

Quality Framework
This data quality framework has been published internationally (Brackstone G., Managing Data Quality in a Statistical Agency, (1999) Survey Methodology, Vol. 25, no. 2, Statistics Canada) and has been recommended by the ANAO as 'better practice' in specifying performance measures (ATO Performance Reporting under the Outcomes and Outputs Framework, Australian Taxation Office, Audit Report No.46 2000-01, pp63-64.) on advice from the ABS Statistical Consultancy Unit.

Return to Survey Contents

---

Relevance

The relevance of statistical information reflects the degree to which it meets the needs of the clients. It is usually described in terms of key user needs, key concepts and classifications used, the scope of the collection and the reference period.

Why do you need to conduct a survey? (B1)

Primary producers, industry organisations and suppliers, and other agricultural service and support industries use agricultural statistics for planning and decision making. The Federal and State Governments make extensive use of agricultural statistics for planning, budgets advisory services and for policies related to marketing agricultural commodities.

The prime objective of the Agricultural Survey is to act as a source of basic agricultural statistics about a wide variety of commodities.

The Agricultural Survey is designed primarily to collect data about commodities (including area and production for crops, number of livestock and area irrigated). Data are also available on the number of producers for each commodity, and on the counts of producers by Australian and New Zealand Standard Industrial Classification (ANZSIC) class.

The Agricultural Survey is also the major source of commodity data input into the calculation of Value of Agricultural Commodities Produced (VACP). VACP is a major output from the survey which is in turn used as an input to the Australian National Accounts.

In addition to the core set of questions the 2006-07 Agricultural Survey contains both internally and externally funded non core questions. The internally funded questions relate to water usage. These questions are being funded by the Environment and Energy Business Statistics Centre. For 2006-07 there will be only one externally user-funded question. This will collect data on new and/or emerging vegetable varieties and several main stream varieties that are not normally collected by the ABS.

How will the survey outputs meet the needs specified in B1? What are the target RSEs for the survey? (B2)

The Agricultural Survey (and five yearly census) is the basic source of Australian agricultural commodity statistics.

The 2006-07 Agricultural Survey will collect data for a wide range of agricultural commodities and land uses including:
Area of holding;
Area of holding by landuse;
Hay and silage (area and production);
Cereal crops (area and production);
Other non-cereal crops (including cotton, sugar cane) (area and production);
Nurseries, cut flowers and cultivated turf (area);
Vegetables harvested for human consumption and for seed (area and production);
Fruit and nut trees, plantation and/or berry fruit (tree number or area and production);
Cattle, sheep, pigs, poultry and other livestock numbers; and
Water use (area and crops irrigated and sources of irrigation water).


Estimates: Final estimates are expected to be available at National, State and Statistical Division (SD). Data will be available by Estimated Value of Agricultural Operations, ANZSIC 06 and the previous ANZSIC 93, plus commodity and specified size ranges (e.g. area of holding size, herd size).

Statistical unit: Data are collected at the agricultural business level.

Scope: As with the 2005-06 Agriculture Census, the scope of the 2006-07 Agricultural Survey includes units flagged as having both agricultural activity and an Estimated Value of Agricultural Operations (EVAO) of $5,000 plus or a Standardised BAS Total Sales (SBTS) of $5000 plus. SBTS is an 'EVAO equivalent' and is calculated by applying ANZSIC based standardisation factors to BAS Total sales for ATOMP units. ABNs identified through ABS frame maintenance activities as not operating an agricultural property are excluded from scope. The number identified on the ABR-based frame (190,000) excludes known duplicate/multiple ABNs from related entity units (a related entity group is a group of ABNs which could all report the same activity on the same parcels of land leading to duplication of reporting) and also ABNs identified through ABS frame maintenance activities as not operating an agricultural property.

Sample size: The 2006-07 Agriculture Survey is expected to increase to around 35,000 units (a sampling fraction of about 21%). The previous Agriculture Survey (2004-05) included around 31,000 units (a sampling fraction of around 24%).

Target RSEs for the survey: RSE constraints on the design variables will be in the order of:
3% for Australian level constraints;
7% for State level constraints; and
15% for Statistical Division and NHT2 constraints.

These constraints were set at 2.5%, 6% and 12% respectively for the 2004-05 reference year and earlier survey years.

Response rate: The 2006-07 Agricultural Survey budget allocation provides for an overall response rate of 80%, which is equivalent to that achieved in previous surveys.

Please specify the principal outputs or data items to be produced. (D1)

Main data items produced
Estimates will be aggregated into various statistical tables/datasets stored on the ABS Information Warehouse. Datasets will include:
Area and production data for all commodities at the national, state and Statistical Division levels;
Industries by estimated value of agricultural operation;
Industries by area of holding;
Industries by Australian Standard Geographical Classification; and
Commodities by yields.

Publications
Principal Agricultural Commodities, Australia, Preliminary (ABS Cat. No. 7111.0). This release contains area and production data, plus comparative data for earlier years, for key commodities at the national and state levels only. Key commodities released in the preliminary estimates include:
Principle crops (wheat, barley, grain sorghum and oats)
Livestock (Milk cattle, meat cattle, sheep and pigs)

Agricultural Commodities, Australia (ABS Cat. No. 7121.0). This release will contain final estimates for area and production of the main commodities collected in the 2006–07 Agricultural Survey. It will contain detailed statistics on broadacre crops, fruit and vegetables as well as the number of livestock and livestock products. Also includes details of industry structure including counts of agricultural businesses by ANZSIC 2006 and size (physical and economic).

ABS website releases
Preliminary estimates are expected to be released November 2007. This release will feature the preliminary commodity set released at the national and state level.

Final estimates at the Australian, state and Statistical Division, are expected to be released on the ABS website in June 2008. This release is expected to contain final estimates for area and production of the main commodities.

Customised output
In addition to the suite of standard publications customised data output such as cross classified and small area data will be available through the various ABS information consultancies units. Standard data will be made available to information consultancy staff in each state via the datasets contained in the ABS Information Warehouse while specialised, cross classified and other custom generated data will be provided by staff in the Clients Services Section of the Agriculture Business Statistics Centre in Hobart.

What statistical data analysis techniques will be performed? (I5)

Comparison with historical data, other collections and other sources

The information that is considered when analysing results include: Survey edits, Historical results, Other collections, Other sources.
Collections: Livestock Slaughtering.
Sources: Industry sources (e.g. Grain boards, sugar refineries, cotton gins), Australian Bureau of Agricultural and Resource Economics (ABARE) data, Media database, Department of Primary Industry (DPI).
Other information: Local knowledge.

What consideration has been given to the use of standards? Please specify. (D2)

Standards used in the Agricultural Census include:

• Australian Standard Geographical Classification regions:
  • National;
  • State; and
  • Statistical Division.
• Industry - both ANZSIC 1993 and 2006;
• Size - Area of Holding and estimated value of agricultural operations; and
• Commodity classes:
  • Each commodity is given a unique Commodity Code Number (CCN);
  • Most commodities will have the same CCN which allows for comparison over time.

What is the target population (scope) for the survey? (G1)

The ABR-based register (Inteframe) used by the Agricultural Census contains approximately 190,000 business units flagged as having both agricultural activity and an estimated value of agricultural operations of $5,000 plus or a Standardised BAS Total Sales (SBTS) of $5,000 plus. SBTS is an 'EVAO equivalent' and is calculated by applying ANZSIC based standardisation factors to BAS Total sales for ATOMP units. ABNs identified through ABS frame maintenance activities as not operating an agricultural property are excluded from scope. The number identified on the ABR-based frame (190,000) excludes known duplicate/multiple ABNs from related entity units (a related entity group is a group of ABNs which could all report the same activity on the same parcels of land leading to duplication of reporting) and also ABNs identified through ABS frame maintenance activities as not operating an agricultural property.

How will you obtain the list of businesses (frame) from which the sample is to be selected? (G2)

The frame will be obtained from the March Common Frame extract of Inteframe, which is an ABR-based business register.

What is the type of unit on the frame to be used for the survey (e.g. farm, school, practice, GP, retail business)? (G3)

Agricultural (farm) business

What is (are) the method(s) of collecting the data (eg self-completion, telephone interviewing, face to face interviewing, etc)? Why was it (were they) chosen? (E3)

The Agricultural Survey uses a mail out/mail back paper collection instrument which is completed by the respondent. This method is the most cost-effective means of conducting the census, and allows respondents to provide information at their convenience (within a 14 day period). Please see Section E1 for a copy of the collection instrument.

Three reminder letters will send out to non-responding units.

Any intensive follow-up is conducted by telephone.

A publicity campaign will be conducted prior to the census despatch to maximise response.

To what reference period(s) does the survey refer? (A6)

Financial Year 2006-07

Is the survey to be conducted once only or repeated? If repeated, with what frequency is the survey to be repeated? (A8)

The survey is an ongoing/repeated collection conducted annually with a census to be conducted every fifth year.




Return to Quality Framework



---


Accessibility



The accessibility of statistical information refers to the ease with which it can be referenced by users. It includes the ease with which the existence of information can be ascertained, as well as the suitability of the form or medium through which the information can be accessed.

How will the results of the survey be explained / presented to the major users? (I6)

The following publications will be released:
7111.0 Preliminary estimates of principal agricultural commodities, including statistics on area and production of principal crops, livestock numbers, lambing, intended matings and comparative data for earlier years.

7121.0 Final estimates for the main commodities collected in the Agricultural Survey, including statistics on area and production of principal crops, fruit, vegetables, livestock numbers, lambing, intended matings, beekeeping and comparative data for earlier years.


Unpublished statistics for certain commodities are available. Some unpublished statistics are available at regional level dependent upon associated Relative Standard Errors (RSEs). This information can be provided by staff in the Agriculture Statistics Centre Hobart (03) 6222 5939, Client Servicing staff in each State and via the National Information Referral Service (1300 135 070).

Which agency and area is responsible for the survey? (A2)

Australian Bureau of Statistics (ABS)


Agriculture BSC (project 070)

Agricultural Statistics Branch

Population, Labour, Industry and Environment Statistics Group

Who is the survey manager and principal contact person for survey clearance? (A3)

Survey Manager:
David Brereton
phone (03) 6222 5804
david.brereton@abs.gov.au

Main Contact Officer for the SCH review template:
Sean Geltner
phone (03) 6222 5968
sean.geltner@abs.gov.au

What commitments have been made to preserve the confidentiality of respondents? How will these be implemented? (F3)

Data Suppression

Ministerial Determination (Statistics Determination 19)

Passive confidentiality - if a provider requests (in writing) suppression of his/her data the request is investigated and if justified, data is suppressed.



Return to Quality Framework



---


Timeliness



The timeliness of statistical information refers to the delay between the reference point to which it pertains and the date on which the information becomes available.

To what reference peiod(s) does the survey refer? (A6)

Financial Year 2006-07

What is the timetable for the survey? (B3)

Testing - key dates:
Test form design and approval finalised early January 2007
Test form will be posted out to target population mid January 2007
Close off for returning the testing forms will be late January 2007
Analysis and reporting of the testing will be complete by early February 2007

Final questionnaire - key dates:
Form content is to be finalised late December 2006
Form drafted and finalised early March 2007
Form approval complete by mid April 2007
Form definition and testing (for the OCR scanning system) late April 2007
Form printing from mid April to early May 2007
Forms ready for despatch from June 2007

Data collection dates:
Data collection will commence from 1 July 2007
Data cut-off for preliminary estimates is expected to be the end of September 2007
Data collection will cease from end February 2008 in preparation for final estimates

Data release dates:
Preliminary publication - Principal Agricultural Commodities, Australia, Preliminary (cat. 7111.0) - released November 2007
Final publication - Agricultural Commodities, Australia (cat. 7121.0) - released June 2008

Return to Quality Framework



---


Accuracy



The accuracy of statistical information is the degree to which the information correctly describes the phenomena it was designed to measure. It includes measures of both sampling and non-sampling error.

What is the expected total number of businesses you will attempt to contact (ie total sample size initially drawn)? How has this been set? (H4)

The total number of units to be approached is about 35,000. This number is required to achieve the RSE constraints for the key commodity variables and takes account of the quality issues and budget considerations.

What allowances will be made for potential sources of non-sampling error in estimation? (I4)

Data is subject to sampling error. The range of errors as measured by Relative Standard Error will vary according to differing respondent numbers for various commodities.
Within a single cycle, there may be variation between preliminary and final estimates due to biased early response patterns. For example, late responses from large remote locations in Western Australia can affect the early estimates for such items as Area of Holding (which are not normally released as preliminary estimates) and cattle numbers.

Of the number of businesses reported in H5, how many businesses are expected to respond (ie the responding sample size)? (H6)

The live response rate is expected to be around 80% in survey years.

What allowances have been made for expected non-response and frame deficiencies in the sample design? (H8)

The Agriculture surveys are known to be subject to a non-response bias generated from not having the time nor resources to determine the live or dead status of every selected unit. Agriculture therefore use respondent mean estimation rather than live respondent mean estimation. A subsample of the Ag Survey selections (called the Live Farm Subsample) is followed up with a view to generating an unbiased estimate of the live populations, and using this to adjust for the non-response bias in estimates.

In what ways does the coverage provided by the frame differ from the target population? (G6)

The coverage of the ABR-based frame is expected to be an improvement over the frames used for Agricultural surveys in the past. Of the full coverage that could be obtained by combining the old Ag Survey frame with the Ag part of the ABR, this frame is expected to provide around 94% coverage of estimated value of agricultural operations, a key summary measure in agriculture. This compares favourably with coverage from the old frame of around 91%. These figures have been calculated using the combined results of the 2001/02 Agricultural Survey and the 2001/02 Agricultural Coverage Survey (which surveyed agricultural business/units on the ABR not on the old Agricultural Survey frame). Much of the coverage loss is accounted for by failing the size scope cut-off, by units with cancelled ABNs and by units with no active in-scope CAC roles. Other coverage loss is accounted for no-ABN units which are not in-scope;

The frame is expected to include a large number of out-of-scope units and units that are related to other in-scope units (requiring the identification of duplicates). It is expected that around 15% of the frame will be out-of-scope or a secondary related entity (duplicate) of other in-scope units.

The proportion of units expected to be defunct is expected to be very small, but unknown.

What testing of the questionnaire(s) has (have) taken place? (E5)

The Agriculture Business Statistics Centre conducts a range of testing programs for all agricultural collections. The testing programs ensure new, amended and existing questions efficiently and effectively collect the required data and keep respondent burden to a minimum.

The main objectives of the 2006-07 Agricultural Survey testing program was to test new sharefarmer related questions and an expanded suite of water use questions.

Testing was conducted in January and February 2007. As the proposed 2006-07 Agricultural Survey form is closely based on the 2005-06 Agricultural Census form, with only a small amount of new content, only limited testing was required.

The new sharefarmer questions were proposed for the final live survey form with the aim to reduce respondent burden by better streaming of the effected providers through the survey form. It was hoped that the new questions would also minimise the need for follow-up contact with effected providers.

During testing the new sharefarming questions performed less than satisfactorily. As a result it has been decided that these questions will require further testing/refining before being used.

The expanded suite of water use questions proposed for the final survey form had been used in previous Agricultural Surveys and also in the 2002-03 Water Survey. These questions had a series of minor amendments made to make them easier to understand/answer and to improve the quality of the data.

Testing showed the suite of water use questions to be performing satisfactorily and recommended their inclusion on the 2006-07 Agricultural Survey form.

What consultations have taken place, other than testing, with businesses or business associations regarding availability of data items and data collection methods? (E6)

The development of the survey strategy has been undertaken in consultation with major users of ABS agricultural statistics, including Commonwealth and State departments responsible for agriculture, the Australian Bureau of Agricultural and Resource Economics (ABARE), the National Farmers' Federation (NFF) and other organisations representing major agricultural commodities and industries.

Annual meetings of users are held in each State. A national User Advisory Group consisting of the major State and national government users and peak industry bodies (e.g. NFF) is convened annually.

How many units are there on the frame? (G4)

The frame population (to be taken from the March Common Frame) is expected to be around 190,000 to 200,000 units. For ABNs known to belong to related entity groups, only one ABN will be included in the despatch, with the other units being excluded from the despatch.

What actions have been taken to overcome known frame deficiencies? (G7)

The move to the ABR-based frame for the census is intended to overcome many of our previously known undercoverage issues.

Of the total number of businesses reported in H4, how many businesses are expected to be live and in scope? (H5)

Using the Live Farm Subsample, we would expect about 87% of the population to be live and in-scope. Since the Agricultural Survey sample is weighted towards larger units (which have a higher live and in-scope rate), we would expect around 90% of the units in the sample to be live and in-scope.

What evidence is there that the expected respondents will be representative of non-respondents? (H7)

The IFU process in place generally targets non-responding units according to their industry and size which mitigates the potential for any such potential bias.

Return to Quality Framework



---


Interpretability



The interpretability of statistical information reflects the availability of the supplementary information and metadata necessary to interpret and utilise it appropriately. It includes appropriate presentation of data such that it aids correction interpretation.

How will the results of the survey be explained / presented to the major users? (I6)

The following publications will be released:
7111.0 Preliminary estimates of principal agricultural commodities, including statistics on area and production of principal crops, livestock numbers, lambing, intended matings and comparative data for earlier years.

7121.0 Final estimates for the main commodities collected in the Agricultural Survey, including statistics on area and production of principal crops, fruit, vegetables, livestock numbers, lambing, intended matings, beekeeping and comparative data for earlier years.


Unpublished statistics for certain commodities are available. Some unpublished statistics are available at regional level dependent upon associated Relative Standard Errors (RSEs). This information can be provided by staff in the Agriculture Statistics Centre Hobart (03) 6222 5939, Client Servicing staff in each State and via the National Information Referral Service (1300 135 070).

What sample design will be used, eg, census, stratified simple random sample? (H1)

The Agricultural Survey is usually a stratified random sample of approximately 35,000 units, designed to produce estimates for all agricultural commodities at the Australian, state and broad regional levels (including Statistical Division (SD) and National Heritage Trust Two (NHT2) regions for 2006-07). In particular, the sample is designed to maximise the quality of estimates for major commodities defined in terms of their importance to Value of Agricultural Commodities Produced (VACP).

The Agricultural Survey is a single stage, stratified random sample.

Selections are undertaken by the Agriculture Statistics Centre.

What stratification has been used in the sample design? (H2)

The Agricultural Survey population is stratified by state, sub-state region (SD and NHT2), industry group and size. It includes a completely enumerated (CE) sector of about 3,100 units which either:

Have extremely large Estimated Value of Agricultural Operations (EVAO) or Area of Holding (AOH); or
Are significant contributors to major commodities.

The sub-state regional stratification used is based on Statistical Divisions. Australian Capital Territory and Northern Territory are both fully sampled and have no 'sub-territory' stratification.

The sample is expected to meet Relative Standard Error (RSE) constraints at National, State and regional levels for Area of Holding and major commodities. The list of major commodities is determined separately for each area of interest (e.g. each State has a different set of major commodities) and includes all commodities which make a significant contribution to the Value of Agricultural Commodities Produced of the area. For area of holding and most major commodities, the RSE constraints are around 3% at national level, 7% at State level and 15% at regional level (SD and NHT2).

What allowances have been made for expected non-response and frame deficiencies in the sample design? (H8)

The Agriculture surveys are known to be subject to a non-response bias generated from not having the time nor resources to determine the live or dead status of every selected unit. Agriculture therefore use respondent mean estimation rather than live respondent mean estimation. A subsample of the Ag Survey selections (called the Live Farm Subsample) is followed up with a view to generating an unbiased estimate of the live populations, and using this to adjust for the non-response bias in estimates.

Will outliers be identified, and, if so, how will they be handled? (I2)

Outliers are 'completely enumerated' (i.e. their weight set to 1 and the weights of all other units in the relevant strata are recalculated) for purposes of estimation. The processes used to identify outliers include:

scrutiny of graphical and tabular listings of data relationships;
estimates - to check for unusual estimates;
contribution to variance;
contribution to estimate; and
contribution to movement.

Will data be aggregated into statistical tables, and if so what are the estimation formulae for the principal output data items? (I3)

Number Raised Estimation

Design Based

This collection produces number raised final estimates for all agricultural commodities collected and a number of derived items (aggregates of commodities).
Estimates produced are: Level estimates, Relative Standard Errors of Level estimates, Proportions, Movement of levels, Movement of proportions.
Level of estimates: Australian, State, Statistical Division and Statistical Local Area (census years only).
Estimates used for: Publication, Value of Agricultural Commodities Produced (VACP), Australian Bureau of Statistics DataBase (ABSDB)

What allowances will be made for potential sources of non-sampling error in estimation? (I4)

Data is subject to sampling error. The range of errors as measured by Relative Standard Error will vary according to differing respondent numbers for various commodities.
Within a single cycle, there may be variation between preliminary and final estimates due to biased early response patterns. For example, late responses from large remote locations in Western Australia can affect the early estimates for such items as Area of Holding (which are not normally released as preliminary estimates) and cattle numbers.

What statistical data analysis techniques will be performed? (I5)

Comparison with historical data, other collections and other sources

The information that is considered when analysing results include: Survey edits, Historical results, Other collections, Other sources.
Collections: Livestock Slaughtering.
Sources: Industry sources (e.g. Grain boards, sugar refineries, cotton gins), Australian Bureau of Agricultural and Resource Economics (ABARE) data, Media database, Department of Primary Industry (DPI).
Other information: Local knowledge.


Return to Quality Framework



---


Coherence



The coherence of statistical information reflects the degree to which it can be successfully brought together with other statistical information within a broad analytical framework and over time. The use of concepts, classifications and target populations promotes coherence, as does the use of common methodology across surveys.

Is the survey a new survey or a cycle of an existing repeated survey? (A7)

Cycle of an existing survey.

The current strategy is for a 5 yearly census with sample surveys in intercensal years. The recent cycles have been:

1996-97 census;
1998-99 sample survey;
1999-2000 sample survey;
2000-01 census;
2001-02 to, and including, 2004-05 sample surveys; and
2005-06 census;
2006-07 sample survey.

Is the survey to be conducted once only or repeated? If repeated, with what frequency is the survey to be repeated? (A8)

The survey is an ongoing/repeated collection conducted annually with a census to be conducted every fifth year.

How will the frame be updated for future survey cycles? (G8)

The ABR-based frame is maintained/updated by information supplied by the Australian Taxation Office and feedback from the 2006 Agricultural Census.

What consideration has been given to making data item definitions consistent between repeats of this survey or across other surveys? (D3)

Where possible data items are consistent from cycle to cycle.

When the Agricultural Census moved from the ABS maintained frame to the ABR-based frame for the 2005-06 cycle there was a break in series. To alleviate some problems associated with this break bridged estimates were produced in addition to the standard estimates.

The 2006-07 Agricultural Survey will carry over the majority of data items used in the 2005-06 Agricultural Census.

Whilst there was a break in series for the 2005-06 cycle because of the change in frame, other data definitions were carried across and continue to be used:
Estimated value of agricultural operations scope cut off has remained at $5,000 since 1993-94 collection cycle.
Most core commodities are carried over from cycle to cycle and also from the intercensal Agricultural Survey.
Most commodities will have the same unique Commodity Code Number which allows for comparison over time.

What alternative sources of data are available and what use is made of them? (C1)

Agricultural statistics are collected by numerous organisations including industry groups, and State and Federal Government departments. There are no alternative sources for many of the items that are collected in the Agricultural Census and Surveys and some other collections are based on different units, scope and coverage. This is the only comprehensive commodity survey conducted. However, some commodity data is collected by ABARE.

In what respects are these alternative sources insufficient? (C2)

Reasons that alternative sources are insufficient include:

Limitations on the type of respondent (e.g. ABARE collects for data from a limited range of ANZSIC industry groups/classes). The Agricultural Census ABR-based frame includes all respondents engaged in agricultural activity, with an estimated value of agricultural operations of $5,000 or more, regardless of their ANZSIC industry group/class;
Limitations on the number of commodities covered, e.g. industry or Government surveys that collect data specific to a particular industry. The Agricultural Census aims for a comprehensive set of statistics about all agricultural activity conducted on the agricultural business (unit) surveyed;
The potential for overlap between the Agricultural Census/survey and ABARE has been investigated. The potential savings of approximately 300 respondent hours is minimal (about 1.5 % of total time taken);
Reporting bias is a concern when considering any voluntary survey as an alternative source; and
Alternative data sources may have too small a sample size to enable the provision of data on a finer geographic resolution other than national/state level.

What consideration has been given to the use of standards? Please specify. (D2)

Standards used in the Agricultural Census include:

• Australian Standard Geographical Classification regions:
  • National;
  • State; and
  • Statistical Division.
• Industry - both ANZSIC 1993 and 2006;
• Size - Area of Holding and estimated value of agricultural operations; and
• Commodity classes:
  • Each commodity is given a unique Commodity Code Number (CCN);
  • Most commodities will have the same CCN which allows for comparison over time.

What consideration has been given to working with other agencies to make their data more suited to your needs? (C3)

Consultations with other agencies and user groups is ongoing.

Examples of agencies consulted include:

SCARM Standing Committee of Agricultural Resource Managers
NCASS National Committee on Agricultural Statistics
NFF National Farmers Federation
DAFF Department of Agriculture, Forestry and Fisheries
ABARE Australian Bureau of Agriculture and Resource Economics
BRS Bureau of Rural Science

There have been ongoing discussions between ABS and ABARE regarding collection rationalisation. The Agriculture Business Statistics Centre recognises the need to minimise overlap between Agricultural Census/Surveys and relevant ABARE surveys. As mentioned in Section C2, the potential for overlap between the Agricultural Census/survey and ABARE has been investigated.


Return to Quality Framework