Close

About

This report is the first publication from the Lowy Institute’s Multiculturalism, Identity and Influence Project, funded by the Australian Department of Home Affairs. Responsibility for the views, information or advice expressed in this report are those of the authors, and do not necessarily reflect the views of the Lowy Institute or the Australian government.


Methodology

The Lowy Institute’s Multiculturalism, Identity and Influence Project, conducted two surveys in November 2020 for this report. The first was a poll of Chinese-Australians, while the second was a survey of the broader Australian population. Both surveys were run in parallel.

Chinese-Australian Poll

Being Chinese in Australia: Public Opinion in Chinese Communities reports the results of a national survey of 1040 adults in Australia who self-identify as of Chinese heritage, conducted between 10 and 30 November 2020 in English, Mandarin (Traditional) and Mandarin (Simplified). The survey was conducted by the Social Research Centre using the Life in Australia™ panel—currently the only probability-based online panel in Australia—and via a nonprobability panel (Multicultural Marketing and Management). The order of questions in the questionnaire was different from the order presented in this report.

Chinese-Australians constitute a population that is dispersed and diverse, including recent immigrants from multiple countries with different native languages who may have difficulty completing a public opinion survey in English. Increased attention paid to Chinese-Australians by both the Australian and Chinese political class may have also made some more reluctant to cooperate with a survey request. Collectively, these characteristics present significant challenges to anyone wishing to survey this population.

Given approximately 5.6 per cent of the Australian population identify as of Chinese ancestry in the census, randomly recruiting a sample of 1000 Chinese-Australians would have required interviewing and screening approximately 20,000 Australians. This approach would not have been practical, and so the Social Research Centre combined probability and non-probability responses to build a representative sample. The qualities of a probability-based survey ensure better coverage of the general population, including panellists who cannot complete surveys online, and are thus more accurate than non-probability panels. However, the number of panellists that identified as being of Chinese heritage, and therefore qualified for the Life in Australia™ survey was 173. Thus, to obtain a sample size sufficient for the analysis of subgroups, inclusion of a nonprobability sample was necessary.

Of the 1040 respondents, 96 responses (completion rate of 55.5%) were achieved via the Life in Australia™ probability-based panel. One hundred and seventy-three members were invited to complete the survey. Members of the panel were randomly recruited via their landline or mobile telephone (rather than being self-selected volunteers). The methodology adopted was a mixed-mode approach, including both online and telephone surveys. The survey was offered to the Life in Australia™ panel in English only, and conducted from 10 November to 23 November 2020.

The remaining 944 responses were obtained from the nonprobability Multicultural Research Panel, operated by Multicultural Marketing and Management. Recruitment was conducted by advertising on social media platforms and through the organisation’s network. A total of 1,152 engaged with the survey with 944 responses (completion rate of 81.9%). The survey was conducted from 11 November to 30 November 2020. Completion rates across the language options were 15.1% in English, 73.2% in Mandarin (Simplified) and 11.7% in Mandarin (Traditional). An IP address trigger was included, and this excluded 25 (2.2%) of panel members who were identified as not responding from inside Australia.

The non-probability sample obtained was over-represented in several categories. Compared to the demographic benchmarks from the census 2016 and from Life in Australia™, respondents had higher levels of education, were younger and less likely to be citizens. Inclusion of the Life in Australia™ responses did reduce the bias, but the over-representation of the abovementioned three categories remained. To address the bias, a superpopulation model for weighting was used. This methodology uses a statistical model to describe the target population and produce sample weights, rather than using selection probabilities. The choice of variables to be included in the model was driven by the availability of population benchmarks and the statistical accuracy achieved. Various combinations of variables were tested to align the combined sample as closely as possible to the target population. The final weighting solution selected includes age, citizenship, education, and parents’ country of birth. Given the limitations, the combined panel approach reduced the bias associated with the non-probability sample and improved accuracy of the results, resulting in an overall weighting efficiency of 48.5% and a margin of error of 1.96 with a 95% confidence interval.

Parallel National Poll

The parallel national survey was conducted on 3,029 adults between 9 November and 23 November by the Social Research Centre. The survey was conducted by the Social Research Centre (SRC), using the Life in Australia™ panel – currently the only probability-based online panel in Australia. Members of the panel were randomly recruited via their landline or mobile telephone (rather than being self-selected volunteers) and agreed to provide their contact details to take part in surveys on a regular basis. SRC uses a mixed-mode approach for the panel, including online surveys (89% of respondents) and computer-assisted telephone interviewing (11% of respondents) to provide coverage of the offline population (households without internet access).


Acknowledgements

Several questions in this report were modelled on those developed by other polling organisations, including the Pew Research Centre, Australian Election Study, Omnipoll, Scanlon Foundation, Ipsos Mori, Chinese-Language Digital/Social Media in Australia: Rethinking Soft Power project, Essential and the Chicago Council on Global Affairs. Fieldwork was managed by Karly Day of the Social Research Centre. Benjamin Phillips, Dina Baz and Sebastian Misson of the Social Research Centre provided design and weighting advice. John Davis of OmniPoll provided independent consulting and reviewed the questionnaires and report. Hannah Leser, intern in the Lowy Institute’s Public Opinion and Foreign Policy program, provided background research and assisted in the development of the questionnaire. Li Xin translated all the surveys. Brody Smith and Stephen Hutchings at the Lowy Institute designed the interactive website. Clare Caldwell edited the report and website. The authors would also like to thank Osmond Chiu, Andrew Chubb, Sheena Greitens, Jieh-yung Lo, Fran Martin, Richard McGregor, Adam Ni, Alex Oliver and Matthew Schrader for their useful contributions.