Using the Expanded CURF

About the Expanded CURF

The data included in the May 2018 Employee Earnings and Hours (EEH) Expanded CURF is released under the provisions of the Census and Statistics Act 1905. This legislation allows the Australian Statistician to release unit record data, or microdata, provided this is done “in a manner that is not likely to enable the identification of a particular person or organisation to which it relates”.

Steps to confidentialise the datasets made available on the Expanded CURF are undertaken in such a way as to ensure the integrity of the datasets and optimise their content, while maintaining the confidentiality of respondents. Intending purchasers should ensure that the data they require, at the level of detail they need are available on the CURF. The Data Items document on the Summary tab contains information about the list of data items, which is available as an Excel spreadsheet on the Downloads tab.

Identifiers

Each employee record has a unique person identifier - ABSPID.

File contents

The May 2018 EEH Expanded CURF can be accessed via the DataLab and is available in SAS, SPSS and STATA formats. The Expanded CURF comprises the following files:

Data files

• SAS FILE: EEH18E.SAS7BDAT contains the CURF data in SAS format
• SPSS FILE: EEH18E.SAV contains the CURF data in SPSS format
• STATA FILE: EEH18E.DTA contains the CURF data in STATA format

Data item file

• DATA ITEMS: contains all the data items, including details of categories and code values, that are available on the Expanded CURF.

Use of weights

The survey was conducted on a sample of employees from a sample of employers in Australia, and as such users need to take this into account when deriving estimates from the Expanded CURF. Each employee record contains a weight (FINPRSWT) and this weight indicates how many employees in the survey scope population are represented by this employee. Where estimates are derived from the Expanded CURF, it is essential they are calculated using the weights.

An employee's chance of selection in the survey varied considerably, depending on their employer's state, sector, industry and size. If an employee's survey weight is ignored, then no account will be taken of the employee's chance of selection, and the resulting estimates may be biased.

A number of the weights in the Expanded CURF have been slightly modified from the original survey weights for confidentiality reasons. This reweighting process has not resulted in significant changes to the estimates and the statistical validity of the Expanded CURF is not affected.

Earnings

Weekly earnings data items have been perturbed and are expressed as continuous data items (in whole dollars only) on the Expanded CURF. Perturbation is a process of slightly altering the reported values to prevent identification of respondents. The distribution of values is not changed significantly through perturbation and the statistical validity of aggregate data is not affected.

Hours paid for

Hours paid for data were requested for all employees. However, hours paid for could not be provided for Managerial employees where there was no relationship between earnings and hours. As a result, data for hours paid for have only been produced for Managerial employees with a link between earnings and hours.

Reconciliation of Expanded CURF with previously published data

Steps to confidentialise the data made available on the Expanded CURF are taken in such a way as to maximise the content of the file while maintaining the confidentiality of respondents. The steps taken to preserve confidentiality include:

• Omitting the States and Territories data item included in previously published output.
• Reducing the level of detail available for the Employer unit size data item (available on the Expanded CURF in two broad groups).
• Including industry data at the Division (1 digit) level only.
• Perturbing earnings data and employee record weights.

As a result, it may not be possible to exactly reconcile all statistics produced from the Expanded CURF with previously published statistics. However, these differences are not significant and should not diminish the value of the Expanded CURF in analysis.