TECHNICAL NOTE SAMPLING ERROR
RELIABILITY OF ESTIMATES
1 As the estimates from the SEE are based on information collected from a sample of public sector 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 public sector 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 a sample. Efforts have been made to reduce non-sampling error by careful design of questionnaires and quality control of processing.
2 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 that would have been obtained from a full enumeration (the ‘true value’ ). 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.
3 An example of the use of standard error on levels estimates is as follows. If the estimated number of employees was 1,400,000 with a standard error of 3,000, then there would be about two chances in three that a full enumeration would have given a figure in the range 1,397,000 to 1,403,000 and about nineteen chances in twenty that it would be in the range 1,394,000 to 1,406,000.
4 The following table shows the standard errors for published estimates for states and territories and level of government for 2010-11. Standard errors for other estimates are available on request.
STANDARD ERRORS, Public sector employees and cash wages and salaries |
|
![](/icons/ecblank.gif) | Employees June 2011 | Cash wages and salaries 2010-11 |
![](/icons/ecblank.gif) | Commonwealth government | State government | Local government | Total | Commonwealth government | State government | Local government | Total |
![](/icons/ecblank.gif) | '000 | '000 | '000 | '000 | $m | $m | $m | $m |
|
New South Wales | 1.6 | 1.8 | 2.5 | 1.7 | 125.7 | 155.4 | 129.0 | 121.7 |
Victoria | 1.7 | 6.4 | 2.9 | 6.2 | 115.2 | 121.5 | 123.1 | 113.2 |
Queensland | 2.0 | 2.1 | 1.8 | 1.2 | 134.3 | 153.5 | 84.3 | 110.1 |
South Australia | 1.2 | 1.2 | 1.5 | 0.8 | 70.1 | 76.5 | 86.8 | 36.8 |
Western Australia | 1.5 | 1.9 | 2.3 | 0.7 | 106.8 | 137.4 | 105.7 | 62.3 |
Tasmania | 0.6 | 0.9 | 1.0 | 0.3 | 36.2 | 71.0 | 42.6 | 31.3 |
Northern Territory | 0.1 | 0.3 | - | 0.3 | 11.1 | 8.4 | 0.6 | 10.6 |
Australian Capital Territory | 0.8 | 0.3 | . . | 0.7 | 58.1 | 48.1 | . . | 39.2 |
Australia | 3.8 | 7.0 | 5.5 | 7.1 | 263.8 | 273.4 | 261.7 | 210.1 |
|
. . not applicable |
- nil or rounded to zero (including null cells) |
5 Another measure of the sampling error (for level estimates only) is the relative standard error, which is obtained by expressing the standard error as a percentage of the estimate to which it refers. Estimates with relative standard errors greater than 25% are not considered sufficiently reliable for most purposes and should be used with caution. These estimates with relative standard errors greater than 25%, in both the tables and data cubes, are denoted by the use of asterisks.