Tags
By Faha
The Westminster Parliament election will be held on July 4th. What will be the results in the 18 Northern
Ireland constituencies ? LucidTalk conducted a poll of voters in the past week. Potential voters were asked their constituency of residence and presented with a list of candidates in their constituency and asked who they would vote for. 3,634 responses were received and this was narrowed down to a final weighted sample of 961 voters.
The results from all 18 constituencies were combined and the results are:
Sinn Fein 23.6%
SDLP 12.5%
Aontu 1.0%
PBP 0.7%
Total Nationalist 37.8%
Alliance 17.0%
Green 1.0%
UUP 12.5%
DUP 21.4%
TUV 5.4%
Total Unionist 39.3%
Other/Independent 4.8%
The results for Sinn Fein and Alliance are almost identical to the 2019 Westminster election. The main difference between the actual results in 2019 and the 2024 poll is that the DUP are down 9.2% from the 2019 election (30.6%). This overstates the DUP decline since Alex Easton was the DUP candidate in North Down and is an independent in 2024. Removing his vote from the 2019 results would put the DUP at 29% in 2019, so a decline of 7.6%. Most of that decline are voters who are now supporting the TUV, who did not compete in 2019.
Could the DUP lose any of their current 8 MPs ? I will look at those 8 constituencies. The potential decline in the DUP vote in 2024 is 25% of their 2019 vote. I will assume a uniform swing of 25% in those constituencies and adjust for boundary changes.
East Belfast
After adjusting for boundary changes the DUP vote and Alliance vote in 2019 would have been approximately 47% each. With a uniform swing decline of 25%, the DUP vote in 2024 would be ~36%. While Alliance will lose a few voters to the SDLP and Green Party, the Alliance vote should still exceed 40% so this appears to be a DUP loss.
Lagan Valley
After adjusting for boundary changes the DUP voter would have been 42% in 2019. A uniform swing decline of 25% would decrease that to 31% in 2024. Alliance received 28.8% in in 2019. However, there will be no Sinn Fein candidate in 2024. The Sinn Fein and SDLP vote in 2019, after adjusting for boundary changes, would have been over 4% for each party in 2019. The Sinn Fein vote alone would give Alliance the victory and it is also likely there will be more tactical voting by SDLP voters for Alliance in 2024. The Jeffrey Donaldson scandal will be most prominent in Lagan Valley so I expect an Alliance win.
Strangford
The boundary changes were extensive here. With the new boundaries, the DUP vote of 47.2% in 2019 would have been 40% . A 25% uniform decline would result in a DUP vote of 30% in 2024. The Alliance vote was 28.4% in 2019 (26% to 27% on the new boundaries). With the new boundaries, the nationalist vote would have been 18% in 2019 and in the District Council election of 2023 it was 19%. It would only require 5% of nationalist voters to tactically vote Alliance to defeat the DUP.
East Antrim
The DUP vote in 2019 would have been 42% based on the new boundaries. A uniform swing decline of 25% would reduce that to 31% in 2024. The Alliance vote in 2019 was 27% with the new boundaries. However, there is ~10% to 11%
nationalist vote in the Westminster and District Council elections with the new boundaries. If half of those nationalist voters vote tactically for Alliance, then the DUP could lose.
North Antrim
The DUP vote was 47.4% in 2019 and with a uniform swing of 25% would be 36% in 2024. This appears to be an easy victory for Ian Paisley. However, Jim Allister is the TUV candidate. In the 2022 Assembly election, the DUP received 25.7% and the TUV 21.3%. Probably a DUP victory, but it could be a close election.
South Antrim
In 2019, the DUP vote was 35.3% and the UUP vote was 29.0%. there was much tactical voting for the UUP by Alliance and nationalist voters. The total unionist vote was 10% higher than in the recent Assembly and District Council elections. With a uniform swing of 25%, the DUP vote would be 27% so it appears that the UUP would win. This is not certain, though, since it is unclear if there will be as much tactical voting for the UUP in this election. The SF vote in the 2023 District Council election was almost 22%, so there is little incentive for SF voters to vote UUP when there is a remote chance that SF could win. What if there is tactical voting for SF by the SDLP or nationalist leaning Alliance voters ? In that scenario, the SF vote could be 25%, which would be uncomfortably close for the DUP.
East Londonderry
With the new boundaries the DUP vote would have been 39% in 2019. A uniform swing of 25% would result in 30% in 2024. The combined non unionist vote was 51% on the new boundaries but SF, SDLP and Alliance were all even at ~16% each. SF received 30% in the District Council elections, so they could theoretically defeat the DUP. However, the SF vote in the LucidTalk poll (23.6%) is considerably less than the 30.9% they received the 2023 District Council elections so it is unlikely that SF will be anywhere near 30% here in 2024. If there is tactical voting by SDLP and nationalist leaning Alliance voters then it could approach 30%.
Upper Bann
The DUP received 41% in 2019. A uniform swing does not apply here since the TUV is not competing. The SF vote has not exceeded 30% in Assembly and District Council elections so this should be an easy win for the DUP.
I calculated the Religion background for the election based on the 2021 census. This would be Religion and Religion Brought Up In. This is based on adding in new voters who are now age 18 (15 to 17 in the census) and subtracting 55,000 deaths of older voters. There are also minor adjustments for British military in the census (do not vote in Northern Ireland), emigration and slight undercount of the Catholic population. I only include people who were born in Northern Ireland, the UK or Ireland. I excluded foreign born residents. Many of the foreign born are EU nationals, most who are not eligible to vote in Westminster elections. The non EU foreign nationals on the electoral register are approximately 1% to 2%.
The results for the voting age population are:
Protestant 47%
Catholic 46%
Other/None 7%
The LucidTalk weighted sample was:
Protestant 48.1%
Catholic 44.2%
Other/None 7.5%
The Catholic proportion is lower than the voting age population. However, Protestant voters are older and Catholic voters are younger and older voters are more likely to vote and younger voters less so. The LucidTalk sample appears to match the makeup of actual voters.
The DUP seats most at risk for a loss are East Belfast and Lagan Valley. Strangford and East Antrim are also at risk but less so. North Antrim and South Antrim are also possible losses but much less likely.
There will another LucidTalk poll in another week. Ideally, it would be helpful to have results for each constituency. The difficulty is that there are only approximately 200 voters per constituency that in the initial sample of 3,634
voters. That is too small a number per constituency. It may be possible to increase the numbers per constituency if more voters participate in the poll. Another approach would be to combine the results of the next poll and also include respondents from the current poll who do not participate next week (to avoid duplication). If the number per constituency was in the 300 to 400 range it could provide a statistically valid sample.