|Page tools: Print Page Print All RSS Search this Product|
QUALITY DECLARATION - SUMMARY
The ABS also uses the HES data in the production of the Consumer Price Index and National Accounts.
The HES is conducted every six years. The 2015-16 HES collected information over the period July 2015 to June 2016.
The first results from the 2015-16 survey were released on 13th September 2017, just over one year from the end of data collection. Subsequent outputs, including feature articles, other publications and Confidentialised Unit Record Files (CURF) will be released later in 2017 and early 2018.
In 2015-16, the HES sample size increased from 9,774 households in 2009-10 to 10,046 households in 2015-16. The HES 2015–16 was run in conjunction with the SIH. The expansion of the 2009–10 sample to include an additional selection of metropolitan households whose main source of income was a government pension, benefit and/or allowance was continued, but with a reduction in sample design enabled through improved screening methods (see Explanatory Notes or the User Guide for more information).
Due to field performance not meeting design expectations, the sample in the first half of the financial year was smaller than expected. An additional sample was implemented in the second half of the 2015-16 financial year to bring the total sample in line with the design. Consequently, the survey sample is concentrated in the second half of the year. Weighting benchmarks for each quarter re-balance the sample.
See the User Guide and Explanatory Notes for more information about sampling and weighting.
To address partial non-response, data were imputed for missing fields. The final HES sample includes 3,487 households (35% of households) and 4,762 person records (25% of persons aged 15 years or over) which had at least one imputed value. Imputation flags are provided at the module level in the CURF to allow users to identify and remove records with imputed data. Full record imputation is also used in the HES to impute data for some partially responding persons to minimise sample loss. Full record imputation is only undertaken for households where the main income earners in the household (e.g. both parents in a 'couple with children' household) have adequately completed the HES but others in the household (for example, children of those parents aged 15 years or over) have not. In 2015-16, 170 person level records were fully imputed, representing less than 1% of person records. Of these, expenditure diary information was fully imputed for 57 persons (less than 1% of the sample).
Reliability of the estimates
Estimates produced from the HES are subject to two types of error: non-sampling error; and sampling error.
Non-sampling error can occur in any collection, whether the estimates are derived from a sample or from a complete collection 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 the data.
The estimates are based on a sample of possible observations and are subject to sampling variability. The estimates may therefore differ from the figures that would have been produced if information had been collected for all households. A measure of the sampling error for a given estimate is provided by the standard error, which may be expressed as a percentage of the estimate (relative standard error).
The sample was designed to facilitate analysis at the part of state level with a high level of accuracy for the indicators from the HES, including average weekly household expenditure and financial stress. Due to the collection issues noted above, a higher level of sampling error may be noted in analysis by quarter of collection.
Each cycle of the HES collects comparable information to allow for analysis of changes over time. Various statistics can be utilised to make comparisons between 2015-16 HES data and data from previous HES cycles. These include proportions (e.g. proportional share of expenditure), means and medians.
The ABS seeks to maximise consistency and comparability over time by minimising changes to the survey. Sound survey practice, however, requires ongoing development and maintenance to maintain the integrity of the data and the efficiency of the collection. For instance, when new products or services become available, these are added to the Household Expenditure Classification. Current income, wealth and expenditure standards are available on the ABS website and more information is available in the Explanatory Notes of this publication and the User Guide publication.
The HES provides baseline expenditure information which provides a comparison point for other ABS and external surveys. A comparison between the HES and the Australian System of National Accounts (ASNA) is provided in the Downloads tab on the User Guide (due for release in late 2017).
Expenditure statistics are generally straightforward to interpret. It should be noted, however, that changes in the dollar value of expenditure over time can be influenced by many factors, including inflation, substitution of cheaper products and services and demographic changes (e.g. aging population, smaller households).
Detailed information on the terminology, classifications and other technical aspects associated with the HES can be found in the Explanatory Notes and Glossary included with this publication. The User Guide (cat. no. 6503.0) will be released in October 2017.
Tabulated data and associated relative standard errors are freely available in Excel spreadsheets which can be accessed from the 'Downloads' tab.
A CURF product will be produced from the SIH, subject to the approval of the Australian Statistician. For further details, refer to the Microdata Entry Page on the ABS website. It is expected that the CURF products will be available in late 2017. A Survey TableBuilder product is also planned to be available in late 2017.
Customised data are also available on request. Note that detailed data can be subject to high relative standard errors which in some cases may result in data being regarded as unfit for release. A data item list is available from the 'Downloads' tab of the User Guide for 2015–16 (cat. no. 6503.0).
These documents will be presented in a new window.