QUALITY DECLARATION SUMMARY
The Queensland Interstate Trade (IST) collection is a quarterly survey which collects information about the interstate movement of goods into and out of Queensland from and to other parts of Australia. This movement can be by purchase, sale or transfer. The information sought is the dollar value of goods by commodity classification of the goods. The classification used is based on the Standard International Trade Classification (SITC).
The main purpose of the IST collection is to satisfy the information needs of the Queensland State Government. The Queensland Treasury use the collection data as input into budgetary matters, including the preparation of Queensland State Accounts, and for industry planning from an import replacement perspective. The estimates are also used internally by the ABS' National Accounts Branch for compiling the annual State Accounts. Summary tables are published on the ABS website (ABS Cat No. 8502.3).
From June quarter 2008, the IST collection has been conducted under the Census and Statistics Act. Prior to June quarter 2008, the data for Interstate Trade was collected under the Queensland Act: 'The Statistical Returns Act, 1896'.
The IST estimates are compiled based on data covering the three months of a reference quarter with the statistics released 14 weeks after the end of that quarter.
Estimates of interstate trade are based on information collected from a sample of registered businesses rather than a full enumeration of businesses and are therefore subject to sampling error. That is, they may differ from the figures that would have been produced if the information had been obtained for all registered businesses. The sampling errors are measurable and provide users of the statistics with an indication of the precision of the estimate. The measure used by the ABS for this is the relative standard error (RSE) - the smaller the RSE, the greater the precision of the estimate.
The IST collection is designed to primarily give a suitable level of precision to the estimates for total imports and exports for Interstate trade between Queensland and other Australian states/territories. Estimates at the more detailed commodity levels are subject to higher RSEs and are less reliable. The IST collection has been designed to achieve RSEs in the range 5-15% for total imports and 10-20% for total exports, in the main these RSEs have been achieved. Estimates with a RSE of less than 25% are considered sufficiently reliable for most purposes.
Other inaccuracies, collectively referred to as non-sampling error, can occur in any statistical collection regardless of whether the collection is based on the full enumeration or a sample. For example, the impact of non-response to the statistical collection, errors in reporting by providers, definition or classification difficulties, errors in transcribing and processing data and under-coverage of the frame from which the sample was selected. If these errors are systematic (not random) then the survey results will be distorted in one direction and therefore unrepresentative of the target population. Systematic errors are called bias.
Every effort is made to reduce non-sampling error to a minimum by careful design and testing of the questionnaires, efficient operating procedures and systems, and appropriate methodology.
The ABS attempts to ensure all its economic statistics are coherent and does this by a number of means. The main ones are use of a single register of business units from which to draw its business population frames, the use of a standard set of business units, the use of a standard set of business oriented classifications (industry, commodity, size, geography). The Interstate Trade collection uses all of these, however, unlike other ABS economic data collections, where the data items collected can be compared across collections, the IST data item is unique. This has an impact on the amount of comparison between its estimates and those of other economic collections.
The IST collection underwent a methodology redesign in 2007 to improve the veracity of the estimates it produces and the first estimates on the new basis were published in respect of the June quarter 2008. The new collection, like most ABS economic surveys, now takes its frame from the ABS Business Register (BR), which is primarily based on registrations to the Australian Taxation Office ABN scheme. The frame is updated annually and takes into account new businesses and businesses which have ceased operating, changes in industry and other general business changes.
The scope of the survey is all businesses with State of Operation Queensland and Business Activity Statement (BAS) stratification* turnover of greater than $1,000,000 covering all industries but with a higher proportion sampled in industries considered more likely to contain businesses undertaking interstate trade.
The new collection was designed to achieve a certain sample size of units conducting Interstate Trade, based on the expected proportion of units in the population undertaking interstate trade. As the units not doing interstate trade are identified, they will be removed from the sample. This could mean some units rotate in or out of the sample and may cause some volatility in the estimates produced.
The new design was reviewed in July 2009 (five quarters after implementation) and it was determined that the current design was appropriate, with some reallocation of units to different strata* being the main design change. A further review will occur in 2012 (four years after implementation).
Rotation of units in and out of the collection occurs each year for the September quarter and it is possible that these rotations may cause unrealistic movements (i.e. not reflective of real-world movements in the population). The survey is designed so that each of the sampled units remain in the collection for a maximum of three years; however, large Interstate Trade units which are completely enumerated are included in the collection indefinitely. This means that each September, approximately one third (it can be more or less than one third due to births and deaths of units to and from the frame) of the sampled units are rotated out and randomly replaced with new units from within the same stratum*. There is the possibility that units doing little or no Interstate Trade can be rotated out and randomly replaced with units doing a large amount of Interstate Trade, which can cause unexpected movements in the estimates.
*Stratification is the process where a population is divided into non-overlapping homogeneous groups called strata that together comprise the whole population. This technique is used to achieve the maximum efficiency from sampling the population.
Industry and size variables are used for stratification in the IST collection. The survey uses the Australian and New Zealand Standard Industrial Classification 2006 (ANZSIC06). This classification system identifies and groups all producing units (both goods and services) in Australia into industries to ensure data compatibility. Size is measured by the Total sales as reported on the Business Activity Statement (BAS) of the business reported to the ATO, known on the business register as stratification turnover.