The Australian Bureau of Statistics (ABS) is preparing to deliver the next Census of Population and Housing (the Census) in 2021. BETA has partnered with the ABS to make it easier and more convenient for people to engage with the Census process. For this project, we are testing whether giving people a number of days to respond to a survey, and by presenting this time period in various ways, leads to greater survey commencement than the traditional ‘due date’ approach.
Intervention start and end date: Friday, 11 October 2019 - Tuesday, 30 June 2020
Ethics approval: BETA ETH2019-02
Experimental design including randomisation:
This was a cluster randomised trial. For this trial, clusters were small statistical areas called mesh blocks that consist of approximately 30 dwellings.
Participants were in the Australian statistical areas of Wagga Wagga and surrounds (New South Wales) and areas south of Brisbane (Queensland). Participants were aware they were invited into the Census test, but they were not aware of details of the trial.
The intervention consisted of two versions of the invitation letter delivered by post. ABS managed the printing and delivery of all materials.
Mesh blocks were stratified by: the difficulty of enumeration in the 2016 census, number of dwellings, enumeration mode in 2016 census (drop-off or mail-out), and mesh block category (residential, primary production etc.). Mesh blocks were selected from strata in groups of four and randomly allocated by computer algorithm to a treatment group on a one-to-one ratio.
All responses and data collection is being managed by the ABS. BETA will not have access to any identifiable information.
Intervention(s): The ABS sends letters to citizens inviting them to participate in a census test. For this trial, there were two versions of the invitation letter and
Control condition: N/A
- Primary outcome: proportion of households that commenced the survey prior to October 18 (i.e. up to and including October 17).
- Secondary outcome: proportion of households that completed the survey prior to October 18 (i.e. up to and including October 17).
Expected sample size: 28,544 households, 892 mesh blocks (approx. 446 per treatment group)