A Methodological Summary of 2020 Irish Election Polling

How have market research companies run polls for this election?

On 8th February 2020, Irish voters will cast their ballots in the 2020 General Election. This article will outline the basic method undertaken by the research companies. This information comes from public articles and data tables.

The article shows survey mode, population, weightings, turnout model, and vote intention questions.

Quota samples

No matter how each company finds respondents, the basic method remains the same.

A sample size is set. Demographic population totals create quota targets. If a poll of 1,000 people requires 490 men and 510 women, the survey continues until these quotas are ‘filled’.

The company gathers the survey sample, and applies weights. This is to make the sample look more like the target population of Irish adults.

From the sample of adults, companies derive a sample of voters. Media partners publish the vote intention shares of this weighted sample of voters.

Single transferable votes in multi-member constituencies decide the Dáil Éireann election. Vote intention questions ask about first preferences.

Image for post
Image for post
According to Red C, this is how vote intentions have changed. (Image: Red C)

The Companies

There are six companies which have conducted vote intention polls since January 2019. Three are AIMRO members: Behaviours and Attitudes, Ipsos MRBI, and Red C. Survation and Panelbase are members of the British Polling Council. Ireland Thinks have also

Behaviours and Attitudes

Survey mode: Face-to-face interviews.
Target population: Irish adults (18+).
Weightings: “Known demographic profile of Irish adults”. B&A use demographic quotas of age, gender, and region.
Turnout model: Only those who say they “would definitely vote” are included. An adjustment is made to respondents who recall voting differently in the last election.
Vote intention question:

If there was a General Election tomorrow, to which party or independent candidate would you give your first preference vote?

Undecided voters are excluded. Support for parties is smoothed by “a technical adjustment factor”.

Image for post
Image for post
This is the full note on the ‘adjustment factor’. (Source: Behaviours and Attitudes)

Ipsos MRBI

Survey mode: Face-to-face interviews.
Target population: Irish adults (18+).
Weightings: The Irish Times ‘poll portal’ includes breakdowns by age, gender, social grade, and region. Other breakdowns are also offered.
Turnout model: People who say they are unlikely to vote are excluded. Those who say the outcome “makes no difference” (giving the lowest answer) on a 10-point scale question are also excluded.
Vote intention question:

This is unclear. Party support refers to “first-preference vote percentages in general elections”. Neither the ‘portal’ nor the article on January 20th include the exact question wording. Undecided voters are not included or allocated in the headline estimates.

Ireland Thinks

Survey mode: Telephone (landline and mobile).
Target population: Irish adults (18+).
Weightings: Age, gender, social grade, region, education level.
Turnout model: Weighted to fit the profile of likely voters, based on voter propensity “from actual election data”.
Vote intention question:

If there was a general election tomorrow, which of the following political parties would you give your first preference to?

The methodological statements are based on a poll from April 2017, and may have changed.

Panelbase

Survey mode: Internet panel.
Target population: Irish adults (18+).
Weightings: age, sex, region, household income, and 2016 election vote recall.
Turnout model: “Likely voters”, with undecided voters excluded.
Vote intention question:

Who do you currently intend to vote for with your first preference vote?

The summary is based on Panelbase’s poll from 24th — 31st January 2020, on behalf of The Times Ireland.

Red C

Survey mode: Telephone (landline and mobile).
Target population: Irish adults (18+).
Weightings: “to profile of all adults”. Red C also applied a past vote recall weighting.
Turnout model: Self-reported likelihood to vote, on a 10-point scale question, and recalling voting in the past election. It is unclear how self-reported likelihood question, as the base is adults who “will vote”.
Vote intention question:

If the general election was tomorrow which party or independent candidate do you think you would give your first preference to?

Unsure voters are excluded from the headline estimate. Like B&A, Red C adjust vote intentions based on past vote recall:

A further past vote weighting is included that takes the recall for how people voted in the last election, compares this to the actual results and weights the data between the two.

Survation

Survey mode: Internet panel.
Target population: Irish adults (18+).
Weightings: Age, sex, region, 2016 General Election vote recall,
Turnout model: Turnout factor based on their stated likelihood to vote (0–10). As an example, those who say 7/10 are assigned a score of 0.7.
Vote intention question:

If the general election were tomorrow, for which of the following parties would you vote as your first preference?

Undecided voters were excluded from the published estimates.

This blog looks at the use of statistics in Britain and beyond. It is written by RSS Statistical Ambassador and Chartered Statistician @anthonybmasters.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store