A key element of the research design for the AWRS was determining the unit of analysis for the employer population.
The AWRS surveys enterprises (i.e. the head office and all worksites of the enterprise) rather than workplaces (i.e. a single worksite of an enterprise).
For most enterprises within the scope of the AWRS, the enterprise was defined as the legal entity and it had one Australian Business Number (ABN). Large enterprises with diverse operations and/or multiple business units within the legal entity were treated differently in the AWRS and in some cases a discrete company or business unit was selected to be surveyed rather than the legal business entity.
A similar approach is taken by the ABS to survey very large and diverse enterprises that have business entities, sub entities, or branches within the enterprise group that can report production and employment data for similar economic activities.
Further information about how enterprises were selected to be part of the AWRS is available from the Sample frame page.
In many cases the enterprise will operate from a single workplace/site and so the unit of analysis can be either the enterprise or workplace. However, for enterprises that operate from multiple workplaces, it is important to note the distinction.
In deciding to survey enterprises rather than workplaces, the Commission weighed up benefits and limitations of both options in consultation with the project steering committee.
Key considerations for determining the most appropriate unit of analysis for the AWRS included:
Most linked employee-employer datasets have surveyed the workplace rather than the enterprise.
For example, the Australian Workplace Industrial Relations Survey (AWIRS) in both 1990 and 1995 was conducted at the workplace and a similar approach is undertaken for the Workplace Employment Relations Study (WERS) in the United Kingdom. Comparability to other linked datasets was one of the identified benefits of pursuing a workplace-level approach.
However, the Commission, in conjunction with the steering committee, decided that the AWRS would be conducted at the enterprise rather than workplace level for 2 main reasons:
This contrasted to the AWIRS 1995 which was able to draw on the Australian Business Register (ABR) (maintained by the Australian Bureau of Statistics (ABS)) which at that time contained robust information about the population of Australian workplaces. This had enabled a sample to be drawn and for estimates from the AWIRS to be representative of that population. Such robust data is not presently available.
Data collected from employers via the various questionnaire components can be linked to enable analysis of relationships between workplace relations matters and the operations, structure and financial performance of enterprises.
Data collected from employers can also be linked to data collected from employees to provide unique insight into links between workplace relations policies and procedures and outcomes for employees.
Unlike other surveys of the Australian workforce where survey participants are randomly selected from their household, the AWRS was designed to source employees from their employer so that the data collected about employees can be linked to information about their employer. This is a unique feature of the AWRS.
All employees from enterprises with 21 or fewer employees were invited to participate in the AWRS. A random selection of employees from enterprises with more than 21 employees was invited to participate. Further information about the selection of employees is available from the Sample design page.