APPENDIX 2 COMPARING OTHER ABS DATA
INTRODUCTION
1 This publication presents experimental estimates to the ANZSIC class level in table 3.1. These show the relative performance of each industry class.
2 Other ABS publications also present estimates for specific industries or economic activities for the 2005-06 reference year. These publications can be categorised based on the frequency of the statistical collections that produce them, that is:
Annual:
Electricity, Gas, Water and Sewerage Operations, Australia, 2005-06 (cat. no. 8226.0)
Manufacturing Industry, Australia, 2005-06 (cat. no. 8221.0) (to be released on 10 December 2007)
Mining Operations, Australia, 2005-06 (cat. no. 8415.0)
Irregular - service industries survey (SIS):
Retail and Wholesale Industries, Australia, 2005-06 (cat. no. 8622.0).
3 These publications supplement the Australian Industry summary statistics with a detailed examination of the structure and performance of businesses involved in selected economic activities for the 2005-06 reference year.
ANNUAL COLLECTIONS
4 The annual publications listed above present results from statistical collections that also contribute to the estimates shown in this publication. Hence, the estimates in this publication for the Manufacturing, Mining and Electricity, gas and water supply industries will closely resemble those presented in the publications specific to those industries. The only differences in methodology used to produce the two sets of estimates relate to timing.
5 Because of the different processing timetables of the respective surveys, a later version of the ATO business income tax (BIT) data has been available for use in compiling the estimates in this publication compared to those in the industry-specific publications listed above. As the contribution of BIT data to aggregates for these industries is relatively small (1% or less of Australian sales and service income in 2005-06 for Mining and Electricity, gas and water supply), the effect of the use of different versions of the BIT file is not significant.
6 At the time of compiling these estimates, data from the Manufacturing industry collection had not yet been finalised. Hence the estimates in this publication for Manufacturing are based on an earlier version of the data from the Manufacturing collection compared to that which will be used to produce the estimates to appear in Manufacturing Industry, Australia.
IRREGULAR COLLECTIONS
7 The service industries publication mentioned above presents results of a statistical collection of the Wholesale trade and Retail trade industries. The ABS's program of Service Industries Surveys (SIS) focuses on different industries and economic activities each year.
8 Aspects of this SIS collection differ from those which produce the estimates in this publication, and hence differences in results can arise. These differences are further explained below.
9 The estimates in that publication have been compiled and published using the 2006 edition of the ANZSIC (ANZSIC06), whereas the data in Australian Industry are based on the 1993 edition. This limits the extent to which data can be compared between publications. The industry classification change has resulted in a reduction in scope and size of both the Retail trade and Wholesale trade divisions. This has affected the Retail trade division to a greater degree than it has the Wholesale trade division.
10 One reason that the two sets of estimates vary relates to the use of different industry coding practices. For the Australian Industry publication, businesses are coded to ANZSIC industry classes on the basis of the activity reported to the ATO when they registered for an ABN or, for more complex businesses, on the basis of information reported directly to the ABS. On the other hand, the service industries surveys present estimates for industries based on detailed financial and activity data reported in the survey. There are a number of reasons why a business classified to any given ANZSIC industry on the ABS Business Register (ABSBR) may not have been mainly engaged in activities associated with that industry during the 2005-06 reference year. This may be because of inaccurate or incomplete information at the time the business was registered, or it may be because the business has changed activity, either temporarily or permanently.
11 After allowing for the reduced scope of the Retail trade industry under ANZSIC06, the differences in terms of total expenses and total income at the division level are generally 3% or less. Effects may be greater for individual Wholesale trade or Retail trade industries and/or for other variables.
12 When interpreting any differences in estimates, sampling error should also be taken into account. See Technical Note 2 for further information. The Relative Standard Error (RSE) calculated for an estimate has the effect of widening the range of possible values that the estimate might take, thereby either expanding or reducing the difference between estimates of the same variable produced from different statistical collections. In general, RSEs of estimates from the service industries surveys are typically lower than those from the Economic Activity Survey for the variables presented in table 3.1 at the ANZSIC class level.
Print Page
Print All