• Summary
• Downloads
• Explanatory Notes
• Related Information
• Past Releases

Place of Work (POWP) / Imputation flag for Place of Work (IFPOWP)

This variable records the geographic area in which a person worked in the week before Census night. This variable is applicable to all persons aged 15 years or older who listed their employment status as 'employed'.

How this variable is created

Place of Work is a person variable which is determined from written responses to the 'Business name' and 'Workplace address' questions. Images of these questions are provided below.

The majority of responses for Place of Work were coded automatically (69%). The remaining 31% of responses went through a 'bulk coding' process where groupings of the same business/property were assigned a Place of Work code. 10.6% of all responses were coded this way, and the other 20.4% of responses were coded manually using a workplace index.

Responses are coded to units of geography known as Destination Zones (DZNs). Destination Zone boundaries for 2016 have been designed by the ABS following consultation with each State and Territory Transport Authority (STAs), and cover all of Australia. Destination Zones are aggregates of whole Mesh Blocks (though data is not coded to the Mesh Block level), and aggregate to Statistical Area Level 2 (SA2s). Due to the hierarchical nature of the Australian Statistical Geography Standard (ASGS), place of work data is also available at the Statistical Area Level 3 (SA3), Statistical Area Level 4 (SA4), State/Territory and Greater Capital City Statistical Area levels. Although DZNs do not fit neatly into Local Government Area (LGA) boundaries, a DZN to LGA correspondence was created to allow data to also be released at LGA level. This data is available on request.

In 2011, the place of work geography included extra categories for records that, due to the quality of the response, could not be successfully coded to a DZN (code 9999) or SA2 level ('Capital City' and 'State undefined'). In 2016, these records (and those where no response was given) were subject to a new imputation process to ensure all records received a valid DZN code.

The variable Imputation Flag for Place of Work (IFPOWP) has been introduced to identify the geographic level at which imputation was required. Overall, 90% of place of work data did not require any imputation. Coding process improvements in 2016 ensured a suitable donor pool was available for imputation, reflected in the following figures:

• The number of records that were coded to an SA2 but not to DZN was 3.1% (down from 3.6% in 2011).
• The number of records coded to a Capital City but not to SA2/DZN was 0.8% (down from 1.4% in 2011).
• The number of records coded to a State/Territory but not to Capital City/SA2/DZN was 1.0% (down from 4.4% in 2011).
• Records where no response was provided for workplace address was 5.1% (up from 2.5% in 2011).

1 Workplace address not imputed
2 Workplace address imputed - SA2 stated
3 Workplace address imputed - Capital city stated
4 Workplace address imputed - State/territory stated
5 Workplace address imputed - Not stated
@ Not applicable
V Overseas visitor

This question was first asked in the 1971 Census. Similar questions have been asked in every Census since then. The Place of Work variable was used for the first time in the 2006 Census and replaced the Journey to Work variable which was used prior to that.

Data is available for Destination Zone in TableBuilder Pro. Prior to 2011, DZN data was only available via customised request. Destination Zone boundaries were updated for 2016 and the new Imputation Flag for Place of Work variable was introduced.

Place of Work is a hierarchical classification, ranging from the broadest geographic level (Australia), to the finest level (Destination Zone). The categories in this variable have changed to reflect the updated Australian Statistical Geography Standard (ASGS) used for the 2016 Census. Data is also available for Local Government Areas (LGAs) to assist with time series comparison.

Non-response rate

Unlike some other Census variables, the non-response rate is not affected by persons imputed into dwellings that did not return a Census Form, as Place of Work is only applicable for persons with a labour force status of 'employed'. Where a respondent provided a response to the 'Business name' question but not the 'Workplace address' question, this was not considered to be 'non-response'.

Using the Imputation Flag for Place of Work, the non-response rate for this variable was 5.1% in 2016 (2.5% in 2011). For more information, refer to Understanding Census data quality.

Data usage notes

Comparability of Destination Zones over time

Destination Zones are not necessarily comparable across Censuses, due to significant changes to both the boundary and code structure.

While data for LGAs is available for both 2016 and 2011, there may have been changes to LGA boundaries between 2016 and 2011 which need to be taken into account. More information is available in the Australian Statistical Geography Standard (ASGS): Volume 3 - Non ABS Structures, July 2016.

In 2016, coding indexes were supplied by the ABS based on indexes used for other geographic topics in the Census, supplemented with other information from the State Transport Authority (such as lists of businesses and properties). This was the same approach taken as in 2011, but differed from previous Censuses, where the coding indexes were wholly supplied by the State Transport Authority.

Data for Place of Work, Place of Usual Residence and Method of Travel to Work can be cross-classified to examine movements of people to and from work, to analyse transport patterns and assist in the planning for transport systems. However, the Place of Work data question refers to a person's workplace address 'last' week; Place of Usual Residence refers to where a person usually lives (which may or may not be where they were counted on Census night); and Method of Travel to Work asks where the person travelled to work on the day of the Census. This difference of timeframes can produce outliers in the data due to a variety of legitimate reasons.


	Example 1			Example 2
Scenario	A person spent the previous night in Brisbane with a friend and walked to work in Brisbane City. She then caught a train to her parent's place on the Gold Coast (which she also regarded as her usual place of residence) that evening where she was enumerated.			A person mainly worked in Alice Springs in the week prior to the Census. However the person moved to Sydney in the intervening week and took a ferry to their new place of employment
Area of Enumeration	Gold Coast			Manly
Place of Usual Residence	Gold Coast			Manly
Place of Work	Brisbane City			Alice Springs
Method of Transport to Work	Walked Only			Ferry