fbpx

Long Methodology Disclosure Statement for the SA Senate Poll conducted 13-16 May, 2022

 

The poll was commissioned by Centre Alliance and conducted by Lonergan Research in accordance with the ISO 20252 standard, and in compliance with the Australian Polling Council Quality Mark standards (www.australianpollingcouncil.com). Lonergan Research surveyed 644 South Australians Voters aged 18+ between May 13 and May 16. The study was conducted online amongst members of a permission-based panel. After interviewing, data was cell weighted to the latest population estimates sourced from the Australian Bureau of Statistics (https://www.abs.gov.au/), and rim weighted to the 2019 federal election results amongst SA (https://www.abc.net.au/news/elections/federal/2019/results/party-totals). Questions that were used for weighting include age, gender, location and the Lower House vote in 2019’s federal election. The weighting efficiency was 82.6%, with a 4.2% margin of error

A senate ballot paper based on draw, with the security removed was shown to respondents. The respondents were then asked:

Thinking just about the Senate, or the Upper House (the place that reviews government and legislation).

 

When voting for the Senate, you may vote in one of two ways.

Either

  • Above the line by numbering at least 6 of the “party” boxes above the line, in the order of your choice (with number 1 as your first choice)

Or

  • Below the line by numbering at least 12 of the “candidate” boxes above the line, in the order of your choice (with number 1 as your first choice)

 

Are you more likely to vote…? (Text for those who haven’t voted) / Did you vote…? (Text for those who have voted)

 

Respondents were then informed that for the purpose of the poll, we were only collecting information as to how people will vote above the line.

The first preference vote question was the following: ‘Who is your first choice?’ where the response options were based on the order of the draw. The options were not randomised, and respondents could only select one option.

The following questions, in the same format as the first preference vote question were then asked to find the top 6:

  • ‘Who is your second choice?’: where first choice response was not shown
  • Who is your third choice?’: where first and second choice responses were not shown
  • ‘Who is your fourth choice?’: where first, second and third choice responses were not shown
  • ‘Who is your fifth choice?’: where first, second, third and fourth choice responses were not shown
  • ‘Who is your sixth choice?’: where first, second, third, fourth and fifth choice responses were not shown

The weighted count results from top 6 preference votes were used to predict the 2022 senate seats in SA. Our process mimicked the AEC Senate preference flow process. A senate quota was calculated first, and the senator was only elected if they reach the quota. The party with the lowest remainder in each round was reapportioned with the second preference. The preference flow continued until 6 senate seats were found.