|Page tools: Print Page Print All|
6. Usual residents were those who regarded the dwelling as their own or main home. Others present were considered to be visitors and were not selected to participate in the survey.
7. Most of the survey information was collected from individuals by trained interviewers with extensive experience in conducting household surveys.
8. To ensure that information about the value of contributions to superannuation and the amount of superannuation people had accrued was of a high quality, individuals selected to participate in the survey were asked to report this information by referring to a superannuation statement (or payslip where appropriate). If survey respondents were unable to refer to the appropriate records, they were asked whether they would agree to authorise their superannuation fund to provide the information to the ABS. Superannuation funds provided information only for those respondents who agreed to sign a consent form authorising the fund to do so. See Appendix 1 for more information about the collection of financial superannuation information.
Sample size and selection
9. Dwellings were selected at random using a multi-stage area sample of private dwellings. All usual residents aged 15 to 69 years in the selected dwellings were asked to participate in the survey.
10. The initial sample for the survey consisted of approximately 18,000 dwellings. Of the approximately 14,000 households that remained in the survey after sample loss (e.g. households selected in the survey which had no residents in scope for the survey, vacant or derelict buildings, buildings under construction), approximately 12,200 (87%) were fully responding, that is, households where everyone in scope of the survey responded fully to the survey.
11. In total, about 25,100 people responded to the survey. This includes a small number of people (about 250) who were unable, or unwilling, to answer questions about their income but responded to all other parts of the survey.
WEIGHTING AND ESTIMATION
12. Weighting is the process of adjusting results from a sample survey to infer results for the total population. To do this, a 'weight' is allocated to each sample unit. The weight is a value which indicates how many population units are represented by the sample unit.
13. In SEAS there are four main types of 'sample units': persons, households, families and income units. Weights were calculated separately for persons and households, and the household weights were applied to families and income units. Only complete households were given a household weight but all persons, including those who belonged to an incomplete household, were given a person weight. For this reason, an estimate obtained using the person weights will not match the same estimate obtained using household weights. For example, if the estimate of all persons is calculated using person weights it will not match the same estimate calculated by multiplying the number of persons in each household by the household weights.
14. The first step in calculating weights for each person was to assign an initial weight, which was equal to the inverse of the probability of being selected in the survey. For example, if the probability of a person being selected in the survey was 1 in 600, then that person would have an initial weight of 600.
15. Non-response may introduce bias into survey estimates if non-respondents differ systematically from respondents. If a particular variable is found to be associated with non-response, initial weights can be adjusted to reduce bias due to non-response. For SEAS, an adjustment to the initial person weights was undertaken to account for non-response bias associated with particular household types.
16. Following the non-response adjustment, the weights were calibrated to align with independent estimates of the population of interest, referred to as 'benchmarks', in designated sex by age by area of usual residence categories. Weights calibrated against population benchmarks compensate for over- or under-enumeration of particular categories of persons, and ensure that the survey estimates conform to the independently estimated distribution of the population by age, sex and area of usual residence, rather than to the distribution within the sample itself.
17. Survey estimates of counts of persons are obtained by summing the weights of persons with the characteristic of interest.
18. The process for weighting households was similar to the process for weighting persons. Households were first allocated an initial weight equal to the inverse of the probability of selection. Households where all usual residents aged less than 70 years were adjusted for non-response associated with age of reference person, household type and area. All households were then calibrated to align with independent estimates of the number of households.
19. Similarly household weights were used to calculate income unit weights.
RELIABILITY OF ESTIMATES
20. The estimates provided in this publication are subject to sampling and non-sampling error.
21. Sampling error is the difference between the published estimates, derived from a sample of persons, and the value that would have been produced if all persons in scope of the survey had been included. For more information refer to the Technical note.
22. Non-sampling errors may occur in any collection, whether it is based on a sample or a full count such as a census. Sources of non-sampling error include non-response; errors in reporting by respondents or recording of answers by interviewers; and errors in coding and processing data.
23. Non-response occurs when people cannot or will not cooperate, or cannot be contacted. Non-response can affect the reliability of results and can introduce a bias. The magnitude of any bias depends upon the extent of the difference between non-respondents' characteristics and those of people who responded to the survey. See above under 'Non-response adjustment' for more information about methods of adjusting for non-response in the weighting process.
24. The following methods were also adopted to reduce the level and impact of non-response:
25. Every effort was made to reduce other non-sampling error to a minimum by careful design and testing of questionnaires, training of interviewers, asking respondents to refer to records where appropriate, and extensive editing and quality control procedures applied at all stages of data processing.
26. The estimates in this publication are based on information collected over the reference period and due to seasonal effects may not be representative of other time periods in the year.
NOTES ON ESTIMATES
27. Income and superannuation contributions and balances were collected using a number of different reporting periods. For income, the periods included the last financial year for business and property income, and the last pay period for wages and salaries and other sources of private income. For superannuation, the contribution periods ranged from July 1997 to October 2000 and, for superannuation balances, the periods ranged from July 1998 to October 2000.
28. Medians were used instead of means for superannuation values because the distribution of values was found to be highly skewed. Means are not a good measure of central tendency for skewed distributions as they are sensitive to extreme values.
29. Refer to Appendix 1 for more information about the collection and presentation of financial superannuation information.
COMPARABILITY OF DATA
30. Wherever possible, standard question modules were used to ensure that data from the SEAS were comparable with data from other ABS household surveys. However, differences in definitions, the scope of the survey, the sample size and the survey timing should be considered when comparing results of different surveys.
SEAS PRODUCTS AND SERVICES
Employment Arrangements and Superannuation, Australia (cat. no. 6361.0)
31. Publications cover only a selection of key results from the SEAS. A list of the data items included in the survey is available on this site.
32. Customised tables are available from the ABS upon request. The ABS will provide a quotation before proceeding with the request and will provide advice on the interpretation of the statistics.
33. Topics covered by the SEAS include:
34. While the publications generally focus on information about individual persons, statistics are also available from the SEAS in relation to households, families and income units (definitions of these terms are given in the Glossary). Clients who are interested in obtaining data for any of these different units should contact the ABS to discuss their requirements.
RELATED ABS PRODUCTS
35. Readers may also wish to refer to the following ABS publications
Australian National Accounts: Financial Accounts (cat. no. 5232.0)
Directory of Superannuation Related Statistics (cat. no. 1131.0)
Employee Earnings, Benefits and Trade Union Membership (cat. no. 6310.0)
Forms of Employment, Australia (cat. no. 6359.0)
Income Distribution, Australia (cat. no. 6523.0)
Retirement and Retirement Intentions, Australia (cat. no. 6238.0)
Superannuation, Australia (cat. no. 6319.0)
Working Arrangements, Australia (cat. no. 6342.0)
36. Current publications produced by the ABS are listed in the Catalogue of Publications and Products (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.
EFFECTS OF ROUNDING
37. Where estimates have been rounded, discrepancies may occur between sums of the component items and totals.
These documents will be presented in a new window.