|Page tools: Print Page Print All|
3. Smaller businesses believed to be insignificant energy users, particularly in industries dominated by very large energy users, were excluded from coverage. Contributions to estimates at all levels from these businesses are not believed to be statistically significant.
4. Data at national and state levels as well as both industry (ANZSIC classification) and selected activity basis (International Energy Agency (IEA) classification) were collected.
5. The collection consisted of seven tailored mail-out questionnaires. These forms were designed to be industry specific in order to help data providers provide the information in an accurate and timely fashion. Data for central government administration, commonwealth justice and defence (ANZSIC Classes 8111, parts of 8120 and 8200), were collected from an administrative by-product source (the Department of Industry, Tourism and Resources EDGAR database).
6. A census (of approximately 600 businesses) was considered the best means of collecting accurate data from:
7. A sample of approximately 14,800 businesses was drawn from the remaining in-scope industries and weighted to produce fuel and energy use estimates for those industries.
8. When questionnaires were returned to the ABS they were checked for completeness and accuracy and, where possible, follow-up contact was made with data providers to resolve reporting problems. Where contact with the providers could not be made, missing items on incomplete questionnaires were imputed using average data from like businesses for which data were obtained.
9. Businesses were asked to apportion total fuel use across states for four broad categories of fuels (electricity, natural gas, petroleum products, and other fuels). All fuels reported by the business under each category were assigned to states in the same proportions.
10. Data contained in the tables of this publication relate to in-scope businesses operating in Australia at any time during the 12 month period ended June 2002.
BUSINESSES CEASED DURING THE YEAR
11. Contributions from businesses which ceased operations during the reference period have been included in the statistical output.
12. Since some estimates in this publication are based on information obtained from a sample of businesses drawn from the survey population, the estimates are subject to sampling variability. That is, they may differ from figures that would have been obtained if all businesses had been included in the survey. One measure of the likely difference is given by the standard error (SE), which indicates the extent to which an estimate might have varied by chance because only a sample of units was included. There are about two chances in three that a sample estimate will differ by less than one SE from the figure that would have been obtained if a census had been conducted, and approximately 19 chances in 20 that the difference will be less than two SEs.
13. Sampling variability (standard error) can be presented as a relative standard error (RSE) which is obtained by expressing the standard error as a percentage of the estimate to which it refers. The RSE is a useful measure in that it provides an immediate indication of the potential inaccuracy in estimates due to sampling and avoids the need to refer to the size of the estimate. Note that electricity generation for sale, gas and petroleum production data are not subject to sampling variability as all businesses identified in these industries were surveyed.
14. The following table contains estimates of RSEs for the data in tables 5 and 6 of this publication (RSE annotations are described in 'Abbreviations and Symbols', below). The RSEs can be used to give an indication of the likely range of values around an estimate within which the true value lies. For example, the estimate of the amount of petrol used in the construction industry (table 5) is 794 ML; the corresponding estimated RSE is 9%. There are approximately 2 chances in 3 that the true amount of petrol used lies between 91% and 109% of the estimate (i.e. 723 ML and 865 ML). To increase the likelihood to 95% (or to produce 95% confidence bounds) this range should be expanded to 82% to 118% (i.e. 651 ML and 937 ML). Smaller RSEs will result in narrower bounds.
RSE Estimates for end-use fuel consumption, by industry, Australia and by state and territory 2001-02
15. Errors other than those due to sampling may occur because of deficiencies in the list of units from which the sample was selected, non-response, and imperfections in reporting by providers. Inaccuracies of this kind are referred to as non-sampling error, which may occur in any collection, whether it be a census or a sample. Every effort has been made to reduce non-sampling error to a minimum by careful design and testing of questionnaires, operating procedures and systems used to compile the statistics.
16. Individual fuel or energy use items were distributed across states based upon fuel categories (as discussed in Explanatory Notes paragraph 9). For businesses where the true distribution of use across states for a fuel differs greatly from the rest of the item's category, this method may have resulted in some misallocation of the fuel's use to states. Where this phenomenon has been identified in large contributing businesses, direct correction of state splits for specific items has been carried out. Care should be taken in interpreting state tables, particularly for the less commonly used fuels.
17. Caution should be taken when comparing end-use and supply data as presented in this publication. No attempt has been made to balance the figures and scope and coverage issues will only partially explain the discrepancies. Fully balanced supply and use estimates will be published by ABARE and ABS during 2004.
18. ABS publications which may be of interest are outlined below. Please note, older publications may no longer be available through ABS bookshops but are available through ABS libraries. All publications released from 1998 onwards are available on this web site (charges apply).
Energy and Greenhouse Gas Emissions Accounts, Australia 1992-93 to 1997-98 (cat. no. 4604.0) - May 2001.
1986-87 National Energy Survey - Energy Consumption in Industry, Australia (cat. no. 8217.0) - June 1989.
RELEASE OF INFORMATION
19. As well as the statistics included in this publication, other additional unpublished data from the energy survey are available on request. For information on the provision of unpublished data please contact Bob Harrison on 02 6252 7369.
20. ABS publications draw extensively on information provided freely by individuals, businesses, governments and other organisations. Their continued cooperation is very much appreciated; without it, the wide range of statistics published by the ABS would not be available. Information received by the ABS is treated in strict confidence, as required by the Census and Statistics Act 1905.
ABBREVIATIONS AND SYMBOLS
These documents will be presented in a new window.