The earnings in the online database at www.whatjobspay.com.au are from the What Jobs Pay and Job Markets Australia series. The 1st edition of What Jobs Pay appeared in 1990, providing figures for 1988. The 13th edition was published in 2012. No more book editions will be published, because all future updates will be included in this website’s database. The more detailed Job Markets Australia series, which began in 1999 as a CD-ROM database, went online in 2004. The current year’s figures will continue to be published on the specialist www.jobmarkets.com.au website, but the www.whatjobspay.com.au database will also have the Job Markets Australia earnings for the current and earlier years.
In short, www.whatjobspay.com.au combines the current and historical figures from the two series. It includes earnings for additional years and age-groups, where the methodology and available data allowed their calculation, and from late 2019 it also provides earnings figures for missing and earlier years, extending the series to 1978.
All of the earnings on www.whatjobspay.com.au are based on the Australian Bureau of Statistics’ statistics and utilize either its ASCO or ANZSCO Dictionaries for the occupational inclusions. Yorkcross Pty Ltd and Rodney Stinson, the labour market analyst who devised the research methodology, acknowledge the dependence of these earnings on the Bureau’s original statistics and the named dictionaries. It must be emphasised that the original statistics have been transformed through that methodology, which is briefly explained below.
The Bureau’s original statistics are from its Census series and Characteristics of Employment survey (from 2014) and the Census and Employee Earnings, Benefits and Trade Union Membership (before 2014 ), except for the beginning years when the Weekly Earnings of Employees series substituted for the latter. The most recent Census provides income relativities by age-group within each occupation or group, which are established against the All Ages earnings in the appropriate unit group occupation or minor/major group in the other named series in the year of the Census. These relativities are maintained in the inter-Censal years, and then replaced by those from the following Census.
The Bureau supplied to Yorkcross Pty Ltd unpublished Census statistics in specially commissioned cross-tabulated tables having comprehensive disaggregations by age-group and occupation, as well as unpublished annual statistics from the Employee Earnings, Benefits and Trade Union Membership series and its successor, the Characteristics of Employment survey. For the Census years of 1991, 1996, 2001, 2006, 2011, 2016 and 2021 the database’s earnings are the earnings figures supplied by the Bureau. When the earnings figures from the latest Census become available, they replace the earnings previously established for that Census year in the What Jobs Pay and Job Markets earnings series. The Bureau’s Average Weekly Earnings was used in determining estimated earnings by occupation and age for the missing years in the original What Jobs Pay and Job Markets Australia series and for the period 1978 to 1987.
Throughout, the research focus has been on average weekly full-time earnings. This remains so in the online database, owing to the measure’s usefulness for determining soundly-based market-rate earnings by occupation and age for those employed full-time, which is defined as 35 hours a week or more.
When supplying or publishing statistics originating from its surveys, the Bureau indicates which statistics are unreliable for most or any purposes. It does this by advising where the relative standard error (RSE) is between 25% and 50% and where it exceeds 50%, or it simply suppresses them. The methodology for the What Jobs Pay and Job Markets Australia earnings series accepts the Bureau’s advice.
|