This survey was conducted among 1,127 adults between March 15 and March 29, 2021, by the Social Research Centre using Life in Australia™, a nationally representative mixed-mode panel of randomly recruited and selected Australians.
Panelists participated via self-administered web surveys (95.7% of respondents) as well as via interviewer-administered telephone surveys (4.3% of respondents). The margin of sampling error for the full sample of 1,127 respondents is plus or minus 4 percentage points.
Life in Australia™ panelists are initially invited to join via landline or mobile phone (random digit-dialing recruitment) or via their postal address (address-based sampling recruitment) to provide national coverage of all Australian adults who can be reached by phone or at an address. This approach ensures representation from both the online and offline Australian population. Different methodologies are used to encourage response by online and offline panel members, including multiple contact attempts and reminders. All respondents are given a small incentive for joining the panel and another incentive for each survey they complete.
The completion rate for this survey is 82.5% of the 1,366 active panelists who were invited to complete the survey. Accounting for initial recruitment to the panel, the cumulative response rate for this survey is 7.1%.
The questionnaire was developed by Pew Research Center in consultation with the Social Research Centre. The online version of the questionnaire was rigorously tested by the Social Research Centre and Pew Research Center researchers to ensure the logic of the instrument (including skip patterns) and that randomizations were working as intended before launching the survey. The survey was fielded in English.
To ensure high-quality data, the Social Research Centre performed data quality checks to identify any respondents showing patterns of satisficing. This includes checking for respondents who may have sped through sections, straight-lining grids or invalid responses to verbatim questions.
Life in Australia™ uses a model-based weighting approach so that the sample results can be used to draw inferences about the population. (The target population for this survey was non-institutionalized persons ages 18 and older, living in Australia.) Panel and wave weights were used, and the data balanced to match external population benchmarks.
Panel weights are incorporated to account for the probabilities of selection at recruitment. Covariates used in the model to calculate the panel weights include dwelling tenure, gender, highest education, household composition, language spoken at home and telephony status.
Wave weights are incorporated to account for response propensity as well as for panel attrition and the recruitment of new panelists. The model for this survey’s wave weights accounts for geographic location, socioeconomic indexes for areas, gender, citizenship status, language spoken at home, Aboriginal or Torres Strait Islander status, number of adults in the household, employment status, highest education, marital status, telephony status, television viewing and internet usage, life satisfaction and social desirability.
To ensure the final data are representative of adult Australians, the sample was balanced to population benchmarks based on official Australian Bureau of Statistics sources including the 2016 Census, supplemented by recent updates, and the 2017-18 National Health Survey. The benchmark variables used to balance this survey include age, education, dwelling tenure, gender, geographic location and language spoken at home.
In addition to sampling error, one should bear in mind that question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of opinion polls.
