|Page tools: Print Page Print All|
CHAPTER 33. ECONOMIC ACTIVITY SURVEY
33.4 Estimates are also cross classified by industry sub-division, business size, and state / territory.
33.5 Employment is defined in the survey as the number of persons working for businesses during the last pay period ending in June of the financial year, includes working proprietors and partners, employees absent on paid or prepaid leave, employees on workers' compensation who continue to be paid through the payroll, and contract workers paid through the payroll. Excluded are persons paid by commission only, non-salaried directors and volunteers. Self-employed persons such as consultants and contractors are excluded from selected businesses to avoid double counting, as they have their own chance of selection on the survey frame. See Chapter 3 for further information on the concept of employment, and on definitions of employment used in ABS household and business surveys.
33.6 The businesses that contribute to the EASurvey are classified:
33.7 The scope of the EAS includes all employing businesses in the Australian economy except for:
33.8 Government owned or controlled Public Non-Financial Corporations are included.
33.9 Data for the units collected directly are obtained using a mail-out/mail-back collection methodology. This is supplemented by telephone interviewing of selected units. Non-response is kept to a minimum by the use of intensive follow-up. Businesses that do not respond by the due date are initially sent reminder letters and then followed up by telephone.
33.10 The survey reference period is the year ending 30 June. Although financial data estimates relate to the full twelve months, employment estimates relate to the last pay period ending in June of the given year. As a result, estimates of wages and salaries per person employed may be affected by any fluctuations in employment during the reference period.
33.11 In order to minimise the load placed on providers, the strategy for this survey is to use, as much as possible, information sourced from the ATO, thus reducing the size sample for direct collection. The frame (from which the direct collection sample is selected) is stratified using information held on the ABS Business Register. Businesses eligible for selection in the direct collect sample are then selected from the frame using stratified random sampling techniques.
33.12 Businesses are selected to participate in the survey (the direct collect sample) only if they meet two criteria: their turnover exceeds a threshold level and the business is identified as being an employing business (based on ATO information) during the reference period. Turnover thresholds are set for each ANZSIC class so that the contribution of surveyed businesses accounts for 97.5% of total industry class turnover as determined by ATO Business Activity Statement (BAS) data. Businesses which meet neither of these criteria are referred to as 'micro non-employing businesses'. These businesses are not eligible for selection in the sample. For these units, data is modelled based on BAS data sourced from the ATO, then added to the directly collected estimates to produce industry statistics.
33.13 Generalised regression estimation has been used since the 2006–07 survey. This estimation method enables maximum use of observed linear relationships between data directly collected from businesses in the survey and auxiliary information. When the auxiliary information is strongly correlated with data items collected in a survey, the generalised regression estimation methodology will improve the accuracy of the estimates. The auxiliary variables used in this survey were turnover and wages sourced from the BAS data.
33.14 Where possible, historical data are used for imputing non-responding units. Where units have not provided data previously or had not been selected in previous surveys data is imputed based on BAS data sourced from the ATO or Live Respondent Mean is used. Clerical imputation is used where only partial responses are received.
33.15 Survey outliers are treated using the 'surprise outlier' or 'winsorised outlier' technique.
33.16 The estimation procedure includes new business provisions, which adjust estimates to allow for the time lag between businesses commencing operation and their inclusion on the ABS Business Register.
33.17 For further information on estimation methods used in ABS Business Surveys, refer to Chapter 25.
RELIABILITY OF THE ESTIMATES
33.18 Estimates from the survey are subject to both sampling and non-sampling error. The relative standard errors of survey estimates are published in Australian Industry (cat. no. 8155.0).
DATA COMPARABILITY OVER TIME
33.19 In order to provide a high degree of consistency and comparability over time, changes to survey methods, survey concepts, data item definitions, frequency of collection, and time series analysis methods are made as infrequently as possible. Significant changes have included:
The EAS collection was first conducted as a pilot test, using a sample of about 600 units.
The survey was expanded to a full dress rehearsal, with a sample of around 8,000 units. Estimates for 1990–91 were considered for publication, but were held over until 1991–92 estimates were also available. Results published in Business Operations and Industry Performance, Australia (cat. no. 8140.0).
Industry classification changed from ASIC to ANZSIC.
Expansion of the survey form to collect data for the Input-Output Section of the ABS. The expansion involved modifying some existing questions to provide a finer level of detail, and including additional questions. The purpose of these changes was to assist in the compilation of the National Accounts on an Input-Output basis.
Questions about waste management and selected expenses (Input-Output) were expanded into two separate supplementary surveys that accompanied the main EAS survey form. A new, shorter version of the main EAS survey form was also trialled on 400 respondents.
The main EAS survey form (the "core form") was produced in a simplified format, with all questions appearing at the front and Include/Excludes, Notes and other explanatory material at the back. New questions were included on the core form, including whether the organisation was a not for profit organisation. An additional question regarding sponsorship was included on the expenses (Input-Output) form. Selected experimental estimates derived by incorporating business income tax (BIT) data were published for the first time.
Tourism Satellite Accounts forms were collected.
Selected experimental detailed industry estimates derived by incorporating business income tax data were published for the first time.
Data of Donations by business to community projects, and Sponsorship, were collected from a subsample of EAS businesses, as a supplementary survey. Data relating to ANZSIC division A (Agriculture, Forestry and Fishing) are not available from either ABS sources or ABARE.
With the introduction of The New Tax System (TNTS), 2001–02 saw a major change to the EAS collection adopting new statistical infrastructure and a new statistical units model based on the ABN for businesses with simple structures and the TAU for businesses with complex structures. This resulted in a new data series commencing from the 2001-02 reference period. To help bridge the series, key data were presented for 2000-01 on the bases of the old and the new infrastructure.
Results published in Australian Industry (cat. no. 8155.0).
Additional sample and questions included for the (biennial) Information and Communications Technology Industry Surveys. ANZSIC Division A restored to outputs.
Additional sample and questions included to enable compilation of the Tourism Satellite Accounts. Collection of most assets and liabilities data discontinued indefinitely.
Additional sample and questions included for the (biennial) Information and Communications Technology Industry Surveys. Employment data included in outputs as experimental estimates.
Additional sample included to enable estimates produced on the 1993 edition of ANZSIC to be bridged to those for 2006.
ANZSIC06 was introduced together with generalised regression as the new estimation method for sampled units, and supplemented with ATO business activity statement (BAS) for micro non-employing businesses.
Additional questions included to provide more detailed breakdown of sales and service income, purchases and inventories. Data for the private sector of Division O (Public Administration and Safety) now included.
Sample size increased by about one-third (from approximately 15,000 to 20,000) to improve reliability of estimates.
Sample size increased by 200 units to improve reliability of Division B (Mining) estimates.
33.20 For further details contact the Labour Market Statistics Section, on Canberra (02) 6252 7206 or email email@example.com.