|Page tools: Print Page Print All RSS Search this Product|
3 Also excluded are the following persons who are not regarded as employees for the purposes of this survey:
4 A sample of approximately 4,700 employer units is selected from the ABS Business Register to ensure adequate State and industry representation. The sample is updated each quarter to reflect changes in the ABS register of businesses. These changes arise from the emergence of new businesses, takeovers and mergers, changes to industry classification, changes in the number of employees, and businesses which have ceased operations. Such updating of the register can contribute to changes in the estimates of average weekly earnings.
5 The statistical unit for the survey comprises all activities of an employer in a particular State or Territory. Each statistical unit is classified to an industry which reflects the predominant activity of the business in the State or Territory. The statistical units are stratified by State, sector, industry and employment size, and an equal probability sample is selected from each stratum.
6 Since February 1992, survey data for a number of Commonwealth Government and ACT Government departments have been collected electronically. From November 1993, survey data for a number of Northern Territory Government departments and agencies have also been collected electronically.
SURVEY DESIGN CHANGES
7 Prior to the August 1996 survey, in a small number of cases where a business had a significant number of employees in more than one industry, a separate statistical unit was created for each industry. This change in procedures had minimal effect on the average weekly earnings estimates.
8 For some time now the ABS has been using Australian Taxation Office information on Group Employer (GE) registrations to add new businesses to the ABS Business Register. The ABS uses GE information to delete from the Business Register those businesses which have ceased trading, or which are no longer employing staff. The result was that a large number of inactive businesses were deleted from the population for the Survey of Average Weekly Earnings for May 1999. This process is ongoing, but the number of deletions after May 1999 has been and is expected to continue to be small. The impact of these deletions on the estimates for May 1999 was negligible and no adjustments have been made to previous estimates.
9 Improvements to the survey estimation method were introduced from May 1999. Adjustments are made each quarter to account for the time lag between businesses commencing operation and their inclusion on the ABS Business Register.
NOTES ON ESTIMATES
10 Average weekly earnings statistics represent average gross (before tax) earnings of employees and do not relate to average award rates nor to the earnings of the ‘average person’. Estimates of average weekly earnings are derived by dividing estimates of weekly total earnings by estimates of number of employees. Changes in the averages may be affected not only by changes in the level of earnings of employees but also by changes in the overall composition of the employed wage and salary earner segment of the labour force.
11 There are several aspects which can contribute to compositional changes such as variations in the proportions of full-time, part-time, casual and junior employees, variations in the occupational distribution within and across industries, variations in the distribution of employment between industries and variations in the proportion of male and female employees. Such effects may apply differently within the different States and Territories, and over time.
12 Estimates of average weekly earnings are rounded to the nearest 10 cents.
13 Percentage changes are calculated on the actual values and may differ from calculations based on rounded estimates.
COMPARABILITY OF RESULTS
14 It is important to recognise that whilst average employee earnings can be derived from the Survey of Employment and Earnings (SEE) published in Wage and Salary Earners, Australia (Cat. no. 6248.0), these average earnings will not be comparable with those produced from the AWE Survey. The SEE collects the sum of employee earnings paid in all pay periods ending in a calendar quarter including payments to employees which are excluded from the scope of the AWE survey such as retrospective pay, pay in advance, leave loadings and severance, termination and redundancy payments.
15 Whilst the quarterly earnings produced by the SEE are affected by the varying number of pay periods from quarter to quarter the AWE results are unaffected because of the selection of a specified pay period in the middle of a quarter. The SEE earnings data are also more subject to seasonal variations, particularly in the December and March quarters, due to the incidence of holidays taken in January being paid in advance and the earnings being reported in December quarter figures.
16 The AWE series was introduced in August 1981 when it replaced the average weekly earnings series based principally on information from payroll tax returns. Revised estimates of average weekly earnings for the period August 1981 to November 1983 were included in Average Weekly Earnings, States and Australia, March Quarter 1984 (Cat. no. 6302.0) published on 12 July 1984. Users who need a measure of the movement in earnings for a period which spans both the payroll tax based and employer survey series should refer to Table 3 in that publication which presents both series linked to a common index base (August 1981 = 100.0).
RELIABILITY OF ESTIMATES
17 As the estimates in this publication are based on information relating to a sample of employers rather than a full enumeration, they are subject to sampling variability. That is, they may differ from the estimates that would have been produced if the information had been obtained from all employers. This difference, called sampling error, should not be confused with inaccuracy that may occur because of imperfections in reporting by respondents or in processing by the ABS. Such inaccuracy is referred to as non-sampling error and may occur in any enumeration whether it be a full count or sample. Efforts have been made to reduce non-sampling error by careful design of questionnaires, detailed checking of returns and quality control of processing.
18 The sampling error associated with any estimate can be estimated from the sample results. One measure of sampling error is given by the standard error which indicates the degree to which an estimate may vary from the value which would have been obtained from a full enumeration (the ‘true figure’). There are about two chances in three that a sample estimate differs from the true value by less than one standard error, and about nineteen chances in twenty that the difference will be less than two standard errors.
19 An example of the use of a standard error is as follows. If the estimated average earnings was $500.00 with a standard error of $4.00, then there would be about two chances in three that a full enumeration would have given an estimate in the range $496.00 to $504.00 and about nineteen chances in twenty that it would be in the range $492.00 to $508.00.
20 An example of the use of a standard error for a quarterly change estimate is as follows. If the estimate of quarterly change between two quarters was $1.50 and the standard error on this estimate was $1.00, then there would be about two chances in three that a full enumeration would have given a quarterly change estimate in the range +$0.50 to +$2.50 and about nineteen chances in twenty that it would be in the range -$0.50 to +$3.50.
21 Quarterly movements in estimates of average weekly earnings are considered to be statistically significant where they exceed two standard errors.
22 Seasonal adjustment is a means of removing the estimated effects of normal seasonal variation from the series so that the effects of other influences on the series can be more clearly recognised. Seasonal adjustment does not aim to remove the irregular or non-seasonal influences which may be present in any particular week. Influences that are volatile or unsystematic can still make it difficult to interpret the movement of the series even after adjustment for seasonal variation. This means that quarter-to-quarter movements of seasonally adjusted estimates may not be reliable indicators of trend behaviour.
23 The series have been seasonally adjusted from September 1983 and the historical series can be made available on request. The seasonal factors are reviewed annually to take account of each additional year’s original data. The review takes place for the February reference period each year.
24 Details about the method of seasonal adjustment of these series are available on request.
25 The ABS considers that trend estimates provide a more reliable guide to the underlying direction of the data, and are more suitable than either the seasonally adjusted or original estimates for most business decisions and policy advice.
26 The trend estimates in this publication, obtained by dampening out the irregular component from the seasonally adjusted series, are calculated using a centred 7-term Henderson moving average of the seasonally adjusted series. Estimates for the three most recent quarters cannot be calculated using this centred average method; instead an asymmetric average is used. This can lead to revisions in the trend estimates for the last three quarters when data become available for later quarters. Revisions of trend estimates will also occur with revisions to the original data and re-estimation of seasonal adjustment factors.
27 If a series is highly volatile then the trend estimates will be subject to greater revision for the latest few quarters as data become available. However, it is important to note that this does not make the trend series inferior to the seasonally adjusted or original series. In some cases the effect of the irregular component on overall movements will remain in the seasonally adjusted and the original estimates but not in the trend series.
28 For more information, see A Guide to Interpreting Time Series-Monitoring ‘Trends’: an Overview (Cat. no. 1348.0) or contact the Assistant Director, Time Series Analysis on 02 6252 6345.
29 Two feature articles which have appeared in the ABS monthly publication Australian Economic Indicators (Cat. no. 1350.0) may also be of interest:
30 Background information about the average weekly earnings series is provided in Information Paper: New Statistical Series: Employment, Average Weekly Earnings, Job Vacancies and Overtime (Cat. no. 6256.0) published 21 June 1984. Copies are available on request.
31 The following publications contain related information and are available from ABS Bookshops:
32 Current publications produced by the ABS are listed in the Catalogue of Publications and Products, Australia (Cat. no. 1101.0). The ABS also issues, on Tuesdays and Fridays, a Release Advice (Cat. no. 1105.0) which lists publications to be released in the next few days. The Catalogue and Release Advice are available from any ABS office or from the ABS website www.abs.gov.au.
33 In some cases, the ABS can also make available information which is not published. This information may be made available in one or more of the following forms: hardcopy, data tape, floppy disk, and Internet Email. Generally, a charge is made for providing unpublished information. For contact information, please refer to the Inquiries box on the front page of this issue.
These documents will be presented in a new window.