CHAPTER 2 - CENSUS PROCEDURES AND OUTPUT
The 2010-11 Agricultural Census population frame will be based on the Australian Business Register (ABR). Introduced as part of new tax arrangements in July 2000, the ABR presented an opportunity for the collection of statistics from a frame source which contained good coverage of businesses.
All businesses with a turnover of $75,000 or more are required to register on the ABR, and are provided with an Australian Business Number (ABN). Many businesses with a turnover of less than $75,000 also choose to register on the ABR.
The ABR-based agriculture frame consists of all businesses contained on the ABR coded to an agricultural industry, as well as all businesses self-identified as having any agricultural activity.
An ABR based frame has been used as the population frame for the Agricultural Census and Surveys since the 2005-06 reference period.
The 2010-11 Agricultural Census will collect data relating to the period 1 July 2010 to 30 June 2011. Any agricultural commodities produced during that period are in scope of the collection. Other activities undertaken during that period, which are relevant to the collection, such as land use or water use, are also covered. Numbers of livestock and other stock figures will be recorded as at 30 June 2011, being the last day of the reference period.
The 2010-11 Agricultural Census will be dispatched just prior to the end of June 2011, with instructions to return the form within 14 days of receipt.
As with the 2005-06 Agricultural Census and subsequent Surveys, the scope of the 2010-11 Agricultural Census will be all businesses registered for an ABN with an Estimated Value of Agricultural Operations (EVAO), or equivalent, of $5,000 or more.
Sector awareness and assistance
Cooperation and acceptance of the Census by the agriculture sector is essential for the success of the collection. For this reason, an awareness campaign targeting the farming and horticulture communities will be conducted prior to the Census.
The aim of the campaign is to maximise the response rates for the Agricultural Census to ensure the ABS is able to produce quality small area estimates. The campaign will involve press releases, radio interviews and advertisements emphasising the need to complete the form promptly to enable the release of timely data.
The ABS will also be relying on the support of grower organisations and state Departments of Agriculture to encourage farm businesses to provide accurate and timely information.
While the Agricultural Census typically achieves a very high response rate, a small level of non-response is inevitable. Non-response will be accounted for in the processing of Census data, and relative standard errors accompanying published data will provide an indication of the accuracy of the estimates.
Preliminary production estimates collected as part of the 2010-11 Agricultural Census are scheduled for release in late 2011. Final commodity production data, water use and land use data, and value of production data are proposed for release in 2012. The 2012 release will include the release of small area data for the standard output regions identified by the ABS. This will be a significant improvement in the timeliness of the release of data compared with the 2005-06 Agricultural Census.
The ABS also plans to release Gross Value of Irrigated Agricultural Production (GVIAP) at the national and state level by late 2012.
Assuring data quality
Data collected in the Agricultural Census is subject to extensive quality assurance based on logical edits (e.g. if livestock are reported then grazing land should be reported), range edits (e.g. limits are imposed on derived yields to identify potential misreporting of tonnes as kilograms or hectares as acres) and historical edits based on previous data reported for that business. Quality assurance is also performed on aggregated data at the national, state and regional level, investigating any unusual values.
Estimates produced from the Agricultural Census will also be evaluated by a variety of methods. Data confrontation is one method whereby data reported from the Census is compared with data from alternative sources, with allowances made for differences in scope and procedures.
Geographic level of output
The ABS intends to replace the Australian Standard Geographic Classification (ASGC) with the Australian Statistical Geography Standard (ASGS). To be implemented from July 2011, the new standard will be the basis for the spatial presentation of ABS statistics for all collections including the 2011 Census of Population and Housing and the 2010-11 Agricultural Census. It will be based on small area structures referred to as mesh blocks resulting in more stable and consistent spatial units than under the existing ASGC.
Standard output from the 2010-11 Agricultural Census will be produced at the Statistical Local Area (SLA) under the ASGC and the Statistical Area Level 2 (SA2 level) under the ASGS. The release of data under both standards will be consistent with the strategy for the release of the information collected as part of the Census of Population and Housing.
In the new standard, SA2 will be the spatial level most similar to the SLA level in the ASGC. It is therefore anticipated that future comparisons of small area data commencing with the 2015-16 Agricultural Census will be made at SA2 level.
For further information on the Australian Statistical Geography Standard (ASGS), see Information Paper: Outcome from The Review of the Australian Standard Geographical Classification, 2008 (cat. no. 1216.0.55.002).
In addition, standard output according to non-ABS spatial structures will also be produced at Natural Resource Management (NRM) regions and Murray-Darling Basin (MDB) level. As in 2006, it is proposed that Census of Population and Housing data will also be concorded and released on the ABS web site at these levels.
The ABS welcomes advice on other key geographic boundaries which should be considered for inclusion as part of the standard release of 2010-11 Agricultural Census release.
As with the 2005-06 Agricultural Census, the collection of detailed location information from the farm business will support more accurate geo-coding than a survey year. The intention is to code all farm businesses to a mesh block to support the release of data at small, user defined, geographic areas as special data services.
Data access and display
Since the 2005-06 Agricultural Census the range of standard agricultural outputs published by the ABS have been well received by users. However the ABS recognises that users of statistics have increasing expectations around electronic access to data as well as a desire for flexible methods of presentation. To meet these expectations the ABS is investigating various methods of disseminating the Agricultural Census outputs on-line. These methods could include the ability for users to select data for specific regions and items of interest, similar to web-based methods for accessing data from the Census of Population and Housing.
The ABS is therefore interested in obtaining feedback on users’ preferences for accessing Agricultural Census data through the ABS website, including suggestions on content, format and presentation. For instance, data has previously been presented according to region, then by respective commodity/land/water data items, but some users may only be interested in a particular item, or require time series data. The ABS will consider user requirements when developing methods of dissemination via the ABS web site.
Enhancing the Agricultural Census
The Agricultural Census is one of many data collections conducted by the ABS.
As part of the 2005-06 Agricultural Census development, the ABS investigated the public acceptance, and viability, of enhancing the Agricultural Census collection by linking with data from the Census of Population and Housing. While the 2005-06 proposal did not proceed, many users of ABS statistics are still very interested in the possibility of analysing such a rich combined data set.
Users of statistics interested in rural issues, in particular understanding communities and population change in rural Australia and the factors which affect land managers behaviour, are keen to see the linking of Agricultural Census and Census of Population and Housing data proceed. The combined data set would allow the exploration of the connections between farm production and other farm related activities such as water use, and social and economic factors including off-farm work, housing characteristics and cultural information such as country of birth and volunteering.
This project will only proceed if the ABS is confident that there would be no adverse public reaction resulting in an impact on the quality of the Population Census or the Agricultural Census. In addition, external funding to cover the entire cost of the project would need to be obtained.
The cost of undertaking this project would be around $5 million.
ABS is proposing to offer an internet reporting option for the Agricultural Census. This would allow respondents to log onto a secure ABS website to complete their survey form on-line. It is expected that at least 15% of respondents would choose this reporting option.
Chapter 3 outlines the proposed 'essential' core data items to be collected, as well as 'possible' other data items.