Census 2021 – Crunching the numbers

By Faha

More detailed results were released today for the 2021 Northern Ireland Census. These included Religion and Religion Brought Up In, Ethnicity and National Identity. I will present the headline figures and then go into more detail as the headline figures do not tell the whole story.

For current religion the results were:

Catholic                                                      42.31%

Protestant and Other Christian             37.36%

Other                                                             1.50%

None                                                            18.75%    

For Religion Brought Up In the results were

Catholic                                                      45.70%

Protestant and Other Christian             43.48%

Other                                                             1.50%

None                                                              9.32%    

Now, these headline results do not reveal certain underlying factors.

Approximately 8% of the Northern Ireland population is of foreign national origin. 6% were born in

another country and 2% are their children (some of whom are adults now) who were born in the UK. The religious breakdown of this group has not yet been released but the relative percentages for each group are available in the 2011 census and would be similar in 2021. Contrary to what is generally believed Catholics are in the minority in this group and 3.5% of that 8% are Other or None. Over 1% are Eastern Orthodox or Protestants

So, for Northern Ireland the population is approximately:

Native Catholic                                        39%

Native Protestant                                   36%

Native None                                             17%

Foreign National Origin                 8%

What the media focuses on is the increase in the Catholic population. However, it is the marked decline in the Protestant population that is of more significance. In the 2021 census the Protestant and Other Christian population was 711,000. However, once foreign nationals are removed (there are over 5,000 Eastern Orthodox alone) as well as some native denominations that are not obviously Protestant (almost 2,000 stated their religion mixed Catholic-Protestant), the Protestant total is only 685,000. In the 2001 census the Protestant population was 895,000 in an era when there were very few foreign nationals. Those “None” raised Protestant population was approximately 30,000 so the comparable Protestant population was 865,000 versus 685,000 in 2021. That is a decline of 180,000.

That decline is due to 3 factors

#1 Greater number of deaths compared to births (1,000 per year)

#2 Emigration (approximately 2,000 per year)

#3 No longer identifying as Protestant. This latter is the major cause of the decline in the native Protestant population.

Another interesting finding is that the “None” Group (not raised in any religion) is 177,000. This is much higher than what one would expect based on the 2011 census. This group was 101,000 in 2011. In looking at the results of the schools census one would expect this group would have increased to 127,000. So why the additional 50,000 increase ? Approximately 10,000 would be from immigration of foreign nationals with no Religion. Some of the other 40,000 would be those whose parents recorded them in a religion when they were younger (age 8 to 18 in 2011) but who as adults in 2021, did not recall any significant religious connection as a child when they filled out the 2021 census form.

Results were also available for District Councils and the findings for the Protestant % of the population are as follows:

Antrim and Newtownabbey   46.26%           
Armagh City, Banbridge and Craigavon 41.73%  
Belfast  29.74%
Causeway Coast and Glens 45.32%
Derry City and Strabane   21.09%
Fermanagh and Omagh  27.93%
Lisburn and Castlereagh  49.17%
Mid and East Antrim   57.85%
Mid Ulster    27.82%
Newry, Mourne and Down   19.04%
Ards and North Down   55.48%
There is only a Protestant majority in the Ards and North Down and the Mid and East Antrim Councils.

The census results do have major implications for the political future, especially for a Border Poll. There are 4 different groups that will determine the outcome of a Border Poll. I assume that the voting age for a Border Poll would be 16 as it was for the Scotland independence referendum. If such a poll were held in 2024 the voting age population would be approximately:

Native Catholic                                        37.5%

Native Protestant                                    37.5%

Native None                                             17%

Foreign National                                        8%

The results of a Border Poll will be determined by the preferences of the “Native None” group of 17% and the Foreign nationals.

Do we know anything about the preferences of the None Group ?

LucidTalk did a poll just prior to the May 2022 Assembly election and the “None” group voter preferences were:

Nationalist parties             36%

Alliance-Green                    42%

Unionist parties                  22%

The None’s have a low preference for unionist parties and it is possible that a significant majority could be persuaded to vote for a United Ireland.

The foreign nationals are 8% of the voting age population but only 4% of those registered to vote. Approximately 5% are EU nationals and this would be the group most likely to support a United Ireland since a United Ireland would result in the Northern Ireland rejoining the EU.

The census shows that there are 1,470,000 old enough to vote and 1,370,000 were on the May electoral register. Of the 100,000 not on the register at least 60,000 are foreign nationals. Of the 40,000 native population approximately half of those could not be registered due to underlying medical problems (dementia, severe autism, etc. ). I mention this because it is often claimed by certain unionists that there are a large number of unregistered potential unionist voters but they are actually very few of the unregistered.

The election turnout in May 2022 was 870,000. I expect that it could be 400,000 higher for a Border Poll. Those extra voters would be more likely to be younger and thus less likely to be unionist. Others would be foreign nationals who are concerned about Brexit.

The Religion results for the 2021 census are probably very accurate, unlike those for any census from 1971 to 2011. There were only 27,400 who did not fully answer the religion question. Some of those would have been those who designated None but did not list a Religion Brought Up In. There was another estimated 58,700 who did not fill out census forms. There were only 86,000 for which no Religion date was available. In 2011, the number for which no religion data was available was well over 200,000. The census office estimated that those 86,000 in 2021 had a similar religious breakdown to those who actually responded to the census question on religion.

There will be more data released in 2023, including the breakdown of Religion by Age and the breakdown of National Identity by Religion.

As we move across the doorstep into our new Ireland

Well, that’s been a busy few hours since the new release of the 1921 Census data.

For those of you unaware of the figures released today, here are the headline figures.

Catholic: 42.3% (+1.5) Protestant: 37.3% (-4.3) Other: 1.7% (+0.9) No Religion: 19.0% (+2.1)

The figures in brackets are the changes since the previous 2011 Census.

Religion brought up in:

Catholic 45.70% Protestant 43.48% Other 1.50% None 9.32%

National Identity:

British only 31.86% Irish only 29.13%Northern Irish only 19.78%

The above figures are a good starting point and Faha will be publishing a detailed analysis tomorrow. The figures for the above have also been published by Council area which I’m working on now.

What is clear to me is that the days of a unionist dominated statelet are finally over, the demographics are travelling in only one direction. There is a lag in the demographics translating into actual votes due to age considerations of course. What we currently have on that front is a Republican First Minister, a minority of unionists at both MP and MLA level and a complete inability on the part of Unionist political parties to deal with, and communicate those facts to their electorate.

I know that the, not unexpected, strength of the “Other and None” cohorts may be crucial in future elections and Polls but I’m aware that Faha will be addressing this tomorrow.

I’m taking a breath!

Approaching our new dawn

It was an almost impossible to imagine day when we all agreed to the GFA many years ago. I think we’ve moved on since then, just a bit, some of us anyway.

The DUP, as usual, haven’t moved a step. They are still trying to basically ditch that agreement, the one that was voted upon and agreed by all of us in Ireland. Back then they tried to stop it, ditch it and destroy it.

They still are.

The attempts to impose a veto by them are entirely consistent with their views 22 years ago, indeed they’re consistent with their views 322 years ago. That is unfortunate. It is instructive to observe how things have changed since then.

Unionism is no longer a majority position in the north east of Ireland. If intelligence was applied they would engage and discuss with the rest of us how we could advance our future years. The truth is they haven’t a clue how to do do that yet.

Personally, I’m happy to discuss our future with anyone, I hope they would be too.

The awful Brexit debacle in England is exposed to us all now, The North East of Ireland has a unique and exceptional opportunity to escape that, the fact that anyone doubts that is a jaw dropper.

Perhaps we need to consider now the logical, intelligent answer to it all.


Census 2021……. the Results begin to Roll

By Faha

The Census Office released the first results of the 2021 Northern Ireland Census this week. The total population was 1,903,100 which is 92,237 more than the 2011 census.

Age and gender profiles were also released. Details on the population of District Councils and statistics on demographic variables such as ethnicity and religion will be released in October or November.

Who are these additional 92,237 in Northern Ireland? We will know by the end of this year but there are some clues from other administrative data that are available.

It appears that the majority of the increase in the population originates from foreign nationals. If one looks at the 2011 census there were approximately 75,000 foreign nationals with 44,000 from European countries (excluding the Republic of Ireland). NISRA has recorded approximately 80,000 NiNO registrations from 2011 to 2021. Not all of these immigrants would have remained in Northern Ireland. There is one more source that is recent and would be more accurate. Because of Brexit, an EU settlement scheme was implemented for EU nationals who are not UK citizens. There have been over 102,000 EU nationals who have applied for this EU settlement scheme as of September 2021. There are almost 3,000 EU UK citizens on the electoral register and accounting for those not registered and their children I estimate there are 6,000 EU nationals that are UK citizens who did not need to apply to this scheme. There are certainly also some EU non-citizens who never applied to the scheme. There also some who applied who may not be resident in Northern Ireland. Indeed, 6,000 applications were refused, withdrawn or invalid.

I estimatethe 2021 resident EU population at 102,000 which would be an increase of 58,000 compared to 2011. For non-EU foreign nationals there were approximately 17,000 NiNO registrations since 2011 and if 14,000 are still resident then the total of all new foreign nationals would be 72,000. Thus, only 20,000 of the total population increase since 2011 would be native Irish.

There is one major factor that cannot be accounted for and that is the effect of the COVID-19 pandemic that began one year before the census. The full population decline during that time due to emigration appears to be several thousand but the final numbers are not yet available. It may be that the true numbers of new foreign nationals could be less.

Between the census of 2011 and 2021 there were approximately 236,000 births and 156,000 deaths (mainly elderly) for a net increase of 80,000. Now, if 72,000 of the increase of 92,000 was from foreign nationals then only 20,000 would be from the native resident population from the year 2011. That would imply a net emigration of 60,000 since 2011.

NiNO registrations also indicate a net migration of 10,000 from the Republic of Ireland into Northern Ireland since 2011 which would imply an even higher net migration of 70,000 of the native population. I will assume that this net emigration is less at 60,000 to account for a net emigration of 10,000 foreign nationals due to the COVID-19 pandemic.
A net emigration of 60,000 would be approximately 25,000 Catholic, 25,000 Protestant and 10,000 Other/None religious background. It has been known for years that more Protestants than Catholics leave Northern Ireland to attend university elsewhere in the UK and are less likely to return so this estimate is probably accurate.

What are the implications for the religious background of the population? It is difficult to estimate those who will identify as Catholic, Protestant, Other or None for religion because of the increasing secular trend in the previous 10 years. It is possible to estimate the group Religion or Religion Brought Up In.
In the 2011 census the final numbers for Religion or Religion Brought Up In were:
Protestant 875,717
Catholic 817,385
Other 16,592
None 101,169

There was much missing data for the religion question. Almost 160,000 never completed a census form and there were another 80,000 to 90,000 who completed the census who did not provide any or complete religious data. That is 14% of the entire population. However, there was a school census in 2011 that provided the religion data for over 99% of the school population and that data indicated that the Catholic population for all of Northern Ireland was underestimated by over 20,000.
Based on yearly school census data we know that of the 236,000 births since 2011 approximately 85,000 would be Protestant and 120,000 Catholic and 31,000 Other/None. We know from the demographics of the elderly population (where most deaths occur) that of the 156,000 deaths approximately 95,000 would be Protestant and 55,000 Catholic and 6,000 Other/None.
I will now estimate the Religion of “Religion Brought Up In” numbers, I will assume a baseline 2011 Protestant population of 865,000. I subtracted 3,000 Eastern Orthodox and mixed Catholic/Protestant from the 2011 numbers and adjusted another 8,000 downward based on the school census underestimate of Catholics.

My estimate for the Protestant population is:
865,000 2011
Minus 10,000 deaths exceeding births
Minus 25,000 net emigration
830,000 for 2021

For the Catholic population I will start with the official number of 817,385 from 2011. For foreign nationals I am assuming only a slight Catholic majority. While many are from countries with large Catholic populations such as Poland, Lithuania, Slovakia, Hungary, Portugal, Spain and Italy, a significant number (probably 25%) are actually Eastern Orthodox from Romania, Bulgaria, Greece, etc. .
Plus 65,000 births exceeding deaths
Minus 25,000 net emigration
Plus 10,000 net immigration from Republic of Ireland
Plus 33,000 net foreign nationals
900,000 for 2021

With the Catholic undercount in 2011 the actual number could be as high as 920,000.
Of the remaining 152,000 people I estimate 40,000 Other (non-Christian and Eastern Orthodox) and 112,000 None (not raised in any religion).

Of course, the actual numbers who will identify their current religion as Protestant or Catholic will be lower. Indeed, the None Group (overall as well as those not raised in any religion) will be much higher. Both the Protestant and Catholic Religion or Religion Brought Up In will also be lower than my estimates for several reasons. Some of those in 2011 who were included in a religion by their parents when they were children will not identify as being raised in a religion as an adult due to tenuous memories of a religious connection in childhood and may not enter a religion brought up in. Even some adults from 2011 who listed a Religion Brought Up In may not do so in 2021 with the passage of time.
The 2021 census will be much more accurate for all demographic variables that are measured. The census office has estimated that 27,000 did not complete the census based on medical card use by individuals who show up otherwise in the NHS. A separate post census survey estimated that 31,000 did not complete census forms. The last census that was this thorough was in 1961.

When the actual data is released later this year it is the numbers on current religious affiliation that is the most important politically. The “Religion” or “Religion Brought Up In” categories may receive the most attention. However, the None group (which includes those who never had a religion and those who previously were affiliated with a religion but not anymore) is unique politically compared to Catholic or Protestant voters. Recent polls have shown that half the None group vote Alliance-Green with the other half equally divided between nationalist and unionist voters.

The None group is also very divided on a Border Poll, with recent polls showing 40% in favour of a United Ireland and another 40% Undecided. The None group will determine the future of Northern Ireland so the size of this designation in the 2021 census will be helpful in deciding the timing of a Border Poll.

Final Pre-Election Assembly 2022 Polls

By Faha

This Post was due to be published on the evening before 5th May election. Due to a major computer issue it wasn’t. That was my fault and I’m sorry it’s late. However I’m publishing it now as I’m sure many here will be interested to compare with the actual outcome of the election. BD

The Assembly election is tomorrow. Here is my review of the final pre-election polls and Leaders’ Debate. 

LucidTalk conducted their final poll less than 2 weeks before the election. Their results (excluding probable nonvoters and Undecided) were: 

SF                                          26.4% 

SDLP                                       9.5% 

PBP                                         1.8% 

Aontu                                     0.4% 

Total nationalist                 38.1% 

Alliance                                 13.4% 

Green                                      3.5% 

Others                                    2.3% 

UUP                                      13.6% 

DUP                                      19.8% 

TUV                                        9.3% 

Total Unionist                      42.7% 

The majority of the Others are probably unionist (PUP, Alex Easton, Claire Sugden, etc.) as 62% of their 2nd preferences were for unionist parties. 

I estimate the total nationalist at 38.5% and total unionist at 44%. It is unknown what percentage were Undecided but the LucidTalk poll one month earlier in March had 4% Undecided. 

Survation conducted an Assembly poll of a weighted sample of 468 voters, less than 2 weeks before the election date. The results were: 

SF                     20.1% 

SDLP                12.2% 

PBP                     3.0% 

Alliance           12.6% 

UUP                   9.8% 

DUP                  18.2% 

TUV                     6.0% 

Others                8.6% 

Undecided         9.6% 

The margin of error is higher in this poll (4%) due to the smaller sample size. The only clue to the background of the Undecided group is that half revealed that they voted in the EU Referendum and 70% voted Remain and 30% Leave. This indicates the eventual voting preference of this group with 70% choosing mainly non unionist parties. The Green Party is probably the largest among the Other group but no actual data is available. This poll shows that the unionist and nationalist parties are equal but the eventual decisions of the Undecided may put the nationalist parties slightly ahead of the unionist parties. 

The Institute for Irish Studies at Liverpool University also conducted a poll less within 2 weeks of the election. The sample size was 1,270 voters. The results were: 

SF                     26.6% 

SDLP                10.5% 

Alliance           18.2% 

UUP                  12.1% 

DUP                  18.2% 

TUV                     5.7% 

Others                8.7% 

Undecided        ??? 

These are the published results but they do not include Undecided voters. In their March poll the percentage of Undecided among voters who intended to vote was almost 19%. The Undecided percentage is unknown for this recent poll but obviously if it as high as in March the results are more difficult to interpret. As with their March poll no underlying demographics are available so it is unknown if their sample is representative of the Northern Ireland electorate. They are also not member of the British Polling Council (of which Survation, LucidTalk and all major British polling companies belong to). 

The BBC held a Leaders Debate on Tuesday night with the leaders of the 5 main parties. Immediately afterwards LucidTalk conducted a poll of a weighted sample of 731 voters who had watched the debate. Voters were asked to rate the performance of the Leaders on the following scale: 

Very Good/Excellent 

Good/Fairly Good 


Poor/Not Good 

Very Bad/Awful 

The Very Good/Excellent to Good/Fairly Good results for each party leader were 

Colum Eastwood         78% 

Naomi Long                  69% 

Doug Beattie                67% 

Michelle O’Neill           63% 

Jeffrey Donaldson       18% 

The Poor/Not Good to Very Bad/Awful results for each party leader were 

Colum Eastwood         11% 

Naomi Long                  15% 

Doug Beattie                16% 

Michelle O’Neill           19% 

Jeffrey Donaldson       73% 

Clearly Jeffrey Donaldson performed very poorly and made a poor impression on voters who viewed the debate. There was little difference in the perceptions for the other 4 party leaders. It appears that only DUP voters (based on the percentages) thought he had a good debate performance. 

Do these perceptions matter? The same voters were asked if the debate would change their vote in the Assembly election. The results were: 

No, I am still planning to vote for the same political party with my 1st preference or I am still undecided 


Possibly- I may consider changing my 1st preference vote 


I will , or now seriously considering, changing my 1st preference vote, or it has made up my mind who to vote for (presumably undecided voters for the latter) 


Now, 13% of voters who are no longer undecided or will or are seriously considering changing their vote, is a very high percentage. It is unknown to which party they are considering changing there vote to. In view of the very poor performance of Jeffrey Donaldson, it is likely that some are DUP voters who will switch to the UUP or TUV. Since there was very little difference in the perception of performance of the other 4 party leaders, there is unlikely to be any net shift among the SDLP, SF , Alliance and UUP. Overall, the polls are showing approximately 6%+ Undecideds. These will not be going to the DUP but to the other 4 parties. If they divide equally then each would receive 1.5%. 

That would not matter to SF overall, since 1.5% of 6% is 25% of the Undecided and SF is already at 25% of the total vote. It would help the UUP and SDLP since they are only in the 12% range in the election polls, so proportionally their overall vote would go up 0.75%. In the real world they would not go up 0.75% since many voters did not watch or are not aware of the Leaders Debate. However, even a 0.25% or 0.5% increase for the SDLP or UUP could make a difference in a constituency.  

Another important factor is transfers. Both LucidTalk and Institute for Irish Studies (results from March poll only available)  asked voters who their 2nd preference would be. The results were: 


SF voters 

SDLP- 44% 

PBP- 14% 

Alliance-  11% 

Green- 6% 

Nontransferable- 17% 

SDLP voters 

SF- 46% 

Alliance- 16% 

Green- 9% 

PBP- 6% 

Unionist- 8% 

Nontransferable- 7% 

Alliance voters 

SDLP- 33% 

SF- 19% 

UUP- 20% 

Green- 15% 

Nontransferable- 6% 

UUP voters 

DUP- 52% 

Alliance- 13% 


TUV- 10% 

Nontransferable- 6% 

DUP voters 

UUP- 48% 

TUV- 36% 

Nontransferable- 5% 

TUV voters 

DUP- 74% 

UUP- 11% 

Nontransferable- 9% 

The Institute for Irish Studies poll showed the following 2nd preference: 

SF voters 

SDLP- 49.3% 

Alliance- 13.4% 

Green- 10.9% 

PBP-  6% 

Nontransferable- 4.5% 

SDLP voters 

SF- 15.8% 

Alliance- 43.4% 

PBP- 6.6% 

Unionist- 19.7% 

Nontransferable- 6.6% 

Alliance voters 

SDLP- 26.6% 

SF- 9.2% 

Green- 18.3% 

UUP- 16.3% 

Undecided- 17.4%  

Nontransferable- 1% 

UUP voters 

Alliance- 31% 

SDLP-  7% 

DUP- 17% 

TUV- 22% 

Nontransferable- 8% 

DUP voters 

UUP- 49.7% 

TUV- 21.2% 

Nontransferable- 8.6% 

TUV voters 

DUP- 40% 

UUP- 20% 

Undecided- 17.5% 

Nontransferable- 15% 

There are some differences between the 2 polls with LucidTalk showing that SF and SDLP voters are equally likely to transfer to each other whereas Institute for Irish Studies shows SDLP voters much more likely to go to Alliance. Both polls show only ¼ of Alliance voters transferring to unionist parties. The transfer pattern for UUP voters also shows in the Institute for Irish Studies poll that UUP voters are as likely to transfer to non unionist parties compared to other unionist parties. That poll also showed a low rate of TUV transfers to the DUP and many non  transferables. Both these polls do not bode well for the unionist parties’ chances for electing MLAs. 

If one adds all the results together for the 3 polls (a poll of polls) the estimated results are: 

SF                                      24% 

SDLP                                 10% 

PBP                                     2% 

Other Nationalist             1% 

Alliance                            13.5% 

Green                                 3% 

UUP                                  12% 

DUP                                  18% 

TUV                                    7.5% 

Other Unionist                 1.5% 

Undecided                        7% 

I am assuming a 4% Undecided in the latest LucidTalk poll and at least 10% Undecided in the latest Institute for Irish Studies poll. This assumption may or may not be completely accurate. 

There are some differences between the LucidTalk and the other 2 polls. LucidTalk has a much lower percentage of Undecideds. This may be due to the nature of their panel, which is likely to include voters who are more politically engaged (including political party members) than the other 2 polls. For example, in the Survation poll the Undecided were much higher among those with the lowest educational and income levels indicating that working class and  poor voters are more likely to be Undecided. Another difference is the support for unionist parties. LucidTalk is predicating a total unionist vote in the 44% range, which would be higher than the total unionist percentage in the 2019 Westminster election. The other 2 polls are predicting a total unionist vote less than 40%. All the polls are predicting a large increase in the Alliance-Green vote. LucidTalk seems to indicate that all this increased support is coming from nationalist voters while the other 2 polls show that increase coming from both unionist and nationalist voters.   The big unknown is where the votes of the approximately 7% Undecided will go. The poll of the Leaders’ debate indicate they will go to the non unionist parties or UUP. 

The final poll will be on May 5th and will be 100% accurate. 

Assembly Poll January 2022

By Faha

LucidTalk released an opinion poll on voter preferences for the 2022 Assembly election. It was conducted in January.  The panel sample was a weighted sample of 2,925 likely voters (excluded undecided and nonvoters).

Northern Ireland assembly to reopen after three-year suspension | Northern  Ireland | The Guardian

Voter preference is as follows:

Sinn Fein                   25.13%           

SDLP                           10.71%            

PBP                              1.20%             

Aontu                          0.89%             

Total Nationalist       37.93%           

Alliance                       14.02%          

Green                          3.01%  

Other                           1.61%          

DUP                            17.23%            

UUP                            14.02%           

TUV                             12.21%                    

Total Unionist            43.46%      

The Other group would be mainly the smaller unionist parties such as PUP and Conservatives.  

Any changes compared to the Autumn poll are 1% or less, which is well within the margin of error.

The religious background of the weighted sample of 2,925 was

Catholic                  738  (25.23%)

Protestant           1,157 (39.56%)

None                       804  (27.49%)

Other                      225 (7.69%)

My only concern with the poll is the religious background weighting. It is unlikely that the Catholic population of Northern Ireland is only 25.23% as that would be a 20% decline from the 45% in the 2011 census. The actual census results will not be available until Autumn 2022.

Because of political developments in this past week the poll is already outdated. I expect that there will be a new poll no later than March to reflect the collapse of the Assembly and possibly an early election.

Assembly Poll 2021-the Possibilities

By Faha

Lord Ashcroft Polls conducted an extensive survey of Northern Ireland voters in November. I have recently presented the Border Poll results. What has received little publicity is the polling on voter preference for the May 2022 Assembly election. The voters polled were a weighted sample of 3,301 voters with the following demographics:

Protestant               44.32%

Catholic                    38.81%

None                         15.96%

Other                          0.88%

Their polling approach is different in that they asked voters to give their party preference on a scale of 0%

Northern Ireland assembly reopens three years after collapse | Northern  Irish politics | The Guardian

to 100%. If a voter ranked a party in the 90% to 100% range or it was their only choice, that was considered their likely vote. The cross tabs were very useful since they showed what other parties voters preferred and this is useful in determining if a voter could change their mind and switch to another party and also to determine what their 2nd and 3rd preferences would be. The latter is particularly important in assessing who will win the 4th and 5th seats in constituencies.

Overall, these were the results:

SF                      25.02%

SDLP                   9.94%

PBP                     3.76%

Alliance             13.51%

Green                  5.0%

Others                 4.39%

UUP                   10.63%

DUP                   16.17%

TUV                    12.45%

The Others would include Aontu and Independent Republicans on the nationalist side and PUP, UKIP and Conservatives on the unionist side.

The views expressed by the Others on Brexit and a Border Poll indicate that approximately 2% of that group are Aontu or Independent Republicans. So, the breakdown among all voters is:

Unionist parties                41%

Nationalist parties            40.5%

Alliance-Green                  18.5%

Even though the unionist parties are slightly ahead of the nationalist parties the transfer pattern does not bode well for the DUP and TUV and, to a lesser extent, the UUP. If a voter rates a party in the 50% to 90% range, they may give a 2nd or 3rd preference for that party.

For Alliance voters these are the percentages that their voters gave other parties in the 50% to 90% range.

TUV              0%

DUP              5%

UUP            10%

SF                16%

SDLP            29%

PBP              6%

Green          58%

Alliance transfers are 3 times more likely to go to a nationalist party compared to a unionist party. 78% of Alliance voters gave the UUP a 0% rating so there appears to be little affinity of Alliance voters for unionist parties.

For the Green Party the results were:

TUV              0%

DUP              5%

UUP            12%         

SF                11%

SDLP           11%

PBP             25%

Alliance      24%

As with Alliance there is little support for the unionist parties and the UUP received a 0% rating from 83% of Green voters.

For UUP voters these are the percentages that their voters gave other parties in the 50% to 90% range.

TUV           18%

DUP           39%

Alliance     27%

Green        14%

SDLP            5%

This is concerning for the DUP and TUV since it appears that less than 60% of UUP voters would transfer to other unionist parties. Indeed, 65% of UUP voters gave the TUV a 0% rating and 42% gave the DUP a 0% rating.

For DUP voters, 42% give the TUV a rating of 50% or higher and 22% give the UUP a rating of 50% or higher. This indicates that DUP voters would be twice as likely to give a 2nd preference to the TUV instead of to the UUP. It also indicates that there are a significant number of DUP voters that could potentially change their 1st preference to the TUV.

For TUV voters, 35% give the DUP a rating of 50% or higher versus only 12% for the UUP. In previous elections the split was even for TUV transfers and this change is probably due to the fact that many of the new TUV voters are former DUP voters.

Overall, the results show that the unionist vote would be slightly higher than the nationalist vote. However, with the strong preference of Alliance and Green voters to transfer to nationalist parties instead of unionist parties this seems to indicate that there would be more nationalist MLAs elected than unionist MLAs.

Actual constituency polls have not been conducted but LucidTalk is currently polling by constituency in certain constituencies.

Lord Ashcrofts Border Poll- Nov 2021

By Faha

Lord Ashcroft Polls conducted a poll of Northern Ireland voters in November. One of the issues polled was voter preference in a Border Poll. The voters polled were a weighted sample of 3,301 voters with the following demographics:

Protestant               44.32%

Catholic                    38.81%

None                         15.96%

Other                          0.88%

One of the questions asked was:

If there was a referendum (Border Poll) TOMORROW, how would you vote ?

The results were:

Northern Ireland stay in the UK                                    48.65%

Northern Ireland join Republic of Ireland                    40.59%

Undecided                                                                         10.75%

The breakdown by religion was:


Northern Ireland stay in the UK                                       88%

Northern Ireland join Republic of Ireland                        4%

Undecided                                                                             8%


Northern Ireland stay in the UK                                     10%

Northern Ireland join Republic of Ireland                     80%

Undecided                                                                          11%


Northern Ireland stay in the UK                                    35%

Northern Ireland join Republic of Ireland                    47%

Undecided                                                                         18%

           There is very little support for a United Ireland among Protestants and it appears that all those

A Letter To The Undecided College Student.

Protestants who are in favour of a United Ireland or Undecided are Alliance-Green-SDLP voters. It is unlikely that support for a United Ireland among Protestants will ever exceed 10%. It is important to note that of the increasing number of people in the population that have no religion, only 35% would vote to remain in the UK.

The preference by age was:

                                     18-24            25-44             45-64          65+            

Stay in the UK 24% 45% 56% 55%

United Ireland              71%              46%               36%             25%

Undecided                   5%                  8%                 9%             20%

There is a large majority of young voters that prefer a United Ireland and elderly voters are the least likely to support a United Ireland. Every year there will be a new age cohort who will be old enough to vote and only ¼ of those will wish to stay in the UK. That is the future.

The preference by likely 2022 Assembly vote:

                                 DUP-TUV   UUP        Alliance   Green      SF        SDLP

Stay in the UK         100%          94%          20%        21%        2%        12%

United Ireland                                              42%        30%       98%       72%

Undecided                                   6%           38%         50%         1%       16%

It is interesting that support among Alliance-Green voters for remaining in the UK is only in the 20% range with twice that level of support for a United Ireland among Alliance voters. Most of the Undecideds are not unionist voters. Of the 10.75% who are Undecided only 1.6% voted for unionist parties in the 2017 Assembly election. Thus, a Border Poll will be decided by the other 9.2% that voted for nationalist parties or Alliance-Green.

There was also a more indirect question on preference for a United Ireland.

Voters were asked the following questions:

Before Brexit I thought Northern Ireland should be part of the UK, and I still do

 43 %

Before Brexit I thought Northern Ireland should be part of the UK, but now I am less sure


Before Brexit I thought Northern Ireland should be part of a united Ireland, and I still do

31 %

Before Brexit I thought Northern Ireland should be part of the UK, but now I think it should be part of a united Ireland


Brexit appears to have had a significant influence on preference for a United Ireland. Support for a United Ireland was only 31% pre Brexit and Brexit increased this to 44%. There is a core support of 43% for remaining in the UK despite Brexit and another 9% who are undecided about remaining in the UK due to Brexit.

This shift toward a United Ireland due to Brexit occurred mainly among SDLP, Alliance and Green voters.

The extent of current views of Northern Ireland voters on Brexit was revealed in this question:

Do you think the UK’s decision to leave the EU was the right decision as far as Northern Ireland was concerned.

NO – it was not the right decision as far as NI was concerned


YES – it was the right decision as far as NI was concerned

34 %

No Opinion


Only 1/3 of Northern Ireland voters are now in favour of Brexit.

The 2021 census results are not yet available but based on the 2011 census and extrapolating to 2022 it appears that the sample in this poll is overestimating the percentage of voting age Protestants and underestimating the percentage of voting age Catholics.

Assembly Poll – Autumn 2021

By Faha

LucidTalk have released an opinion poll on voter preferences for the 2022 Assembly election. It was conducted in November 2021.  The panel sample was a weighted sample of 3,298 likely voters (excluding undecided and nonvoters).

Voter preference is as follows and changes are from the August poll:

Sinn Fein                   24.20%             – 0.35%

SDLP                           11.61%             – 1.31%

PBP                              1.94%              – 0.52%

Aontu                          0.94%              + 0.36%

Total Nationalist       38.69%            – 1.84%

Alliance                       13.98%            + 1.37%

Green                          2.49%              + 0.37%

DUP                            18.22%             + 5.24%

UUP                            13.80%            – 1.97%

TUV                             11.37%            – 2.45%

Other Unionist            0.81%             – 0.19%

Total Unionist             44.21%          + 0.68%

The major change from the August poll is the increase in the DUP voter preference and the corresponding decline in the UUP and TUV preference. These are the only statistically significant changes since the August poll. The DUP preference has now returned to the level it had in January, before all the leadership turmoil occurred.

Unlike previous polls, there was no breakdown of preference by religion. This may be due to the fact that the census data on religion is more than 10 years old and the 2021 census data will not be available until 2022.

However, there were results presented on changes in voter preference compared to how the respondents voted in the 2017 Assembly election. The information shows the source of the increased Alliance preference. There was a net loss of 31 voters from Unionist parties to Alliance and a net loss of 94 voters from Nationalist parties to Alliance. Thus, there are 3 times as many nationalist voters that have defected to Alliance compared to unionist voters.

For unionist parties Alliance has gained a net of 18 UUP voters and 11 DUP voters.

For nationalist parties Alliance has gained a net of 50 SDLP voters and 42 SF voters.

The changes for the UUP are only 0.5% of all voters and this has implications for the UUP strategy in the 2022 Assembly election. It is unlikely that the UUP will take a more moderate stance on any contentious issue since that 0.5% is not worth trying to win back with more moderate policies. Indeed, doing so would likely result in a greater loss of UUP votes to the DUP or TUV.

The next poll will probably be in February 2022, which is only 3 months before the May election. The only major development that I foresee before then would involve Article 16 of the Northern Ireland Protocol. Depending on what happens with the British negotiations with the EU, the February results may reflect further voter intention changes.

The Setting of our Boundaries

By Faha

The Northern Ireland Boundary Commission released their proposals for the new boundaries of the 18 Westminster constituencies. I will look at the proposals for each constituency and analyze how the changes would influence the results for both Westminster and Assembly elections.

Foyle and East Londonderry

A unionist section of the Slievekirk ward was transferred from West Tyrone to Foyle. The Eglington ward (~55% Catholic) was transferred from Foyle to East Londonderry. The overall demographic change would be neutral for Foyle and East Londonderry

Should the Assembly break the constituency link with Westminster? -  Research Matters

would be 0.5% more Catholic.
There would be no changes in the Westminster election. However, for an Assembly election in East Londonderry the loss of one of the unionist seats is more likely. The combined unionist vote in the 2019 Westminster election was 49.3%, slightly less than 3 quotas. Sinn Fein, Alliance and the SDLP all had a vote from 15.1% to 15.7% with 4.4% for Aontu. Decreasing the total unionist vote by 0.5% would make it more likely that Sinn Fein, Alliance and the SDLP would all win a seat with a net loss of one unionist.

West Tyrone
A unionist section of Slievekirk was transferred to Foyle. The new wards of Pomeroy and Donaghmore were transferred from Mid Ulster and Fermanagh South Tyrone to West Tyrone. Overall, the constituency is 1% more Catholic so any hope of regaining a 2nd unionist seat in an Assembly election is unlikely.

Fermanagh South Tyrone
There were major changes here. The Dungannon town wards were transferred to Mid Ulster and the unionist Killyman ward was transferred to Upper Bann. The old Loughgall, Ballymartrim and Charlemont wards were transferred from Newry and Armagh. The overall change is to increase the Catholic population less than 0.5%.
However, the changes in an election would be more pronounced. The Dungannon wards have a high percentage of Catholic EU nationals who do not vote in Westminster elections whereas there is a high turnout of nationalist voters in the wards added from Newry and Armagh. I estimate that the Sinn Fein vote would increase 1% to 2% in a Westminster election. The prospect of a unionist candidate regaining this seat in a Westminster election appear more remote with these changes.
Mid Ulster gains the Dungannon town wards and loses the Pomeroy ward and the new Donaghmore ward to West Tyrone. Overall, the constituency is 1% less Catholic. It would appear that there would be an increased chance for a 2nd unionist Assembly seat. However, in the 5 years between elections that 1% decrease will be erased by demographic changes.
Upper Bann gains the unionist Killyman ward from Fermanagh South Tyrone and loses the Loughbrickland ward to South Down. It also loses the Aghagallon ward (85% Catholic) and the Magherlin ward (25% Catholic) to Lagan Valley. Overall, it is 1.5% less Catholic. In a 2027 Assembly election that 1.5% decline will be erased by demographic changes over the 5 years between elections.

South Down and Strangford
These 2 constituencies have the greatest changes and much of it involves exchanges of wards between them.
Strangford gains the entire 5 wards of the Downpatrick DEA (only 7% unionist vote in 2019 Council election). It also gains part of the Catholic majority Crossgar ward. The Saintfield and Moneyreagh wards are transferred to Belfast South and the Kilmore and Ballynahinch wards are transferred to South Down. South Down also gains the Gransha ward from Lagan Valley.
The Catholic population to Strangford increases from 17% to 30%. Based on the 2019 Westminster results the total non unionist vote would equal or exceed 50%. This would result in the loss of 2 unionist seats in an Assembly election with the SDLP and Alliance each gaining one.
The Catholic population of South Down would decline 7.5%. Based on the 2019 Westminster results the total unionist vote would increase to 30%, short of 2 quotas. It is likely that the SDLP would lose a seat to Alliance.

North Antrim and East Antrim
The North Antrim wards of Glenravel, Glenwhirry and Slemish are transferred to East Antrim and North Antrim gains the old Glendun ward (93% Catholic). The 3 wards transferred to East Antrim are overall identical in demographics to the remainder of North Antrim. The addition of the Glendun ward to North Antrim increase the Catholic population of North Antrim by almost 1%. The net effect would be a loss of a unionist Assembly seat to Alliance based on the 2019 Westminster election.
South Antrim has only minor changes with the loss of some partial wards that are heavily unionist to East Antrim and the Catholic majority sections of Mallusk ward to North Belfast.
Lagan Valley loses the Gransha ward to South Down and gains the Aghagallon ward (85% Catholic) and Magherlin ward (25% Catholic). Overall, the Catholic population increases by 4%. This would solidify the SDLP seat here.

North Down
Only one minor change here with the addition of the Garnerville ward (similar to the old Belmont ward) from East Belfast.

North Belfast
The unionist Forth River ward was added and the unionist Woodvale and Shankill wards were transferred to West Belfast. The Catholic majority sections of Mallusk ward were added. Overall, the Catholic population increases by over 2%. This will have significant implications for the next Westminster election. When also accounting for demographic changes between 2019 and 2024 it appears that the total unionist vote could drop below 40%. The Sinn Fein vote could reach or exceed 50%. This constituency appears to be lost forever to unionism.

East Belfast
Garnerville ward was transferred to North Down and the Woodstock and old Wynchurch and Hillfoot wards were added from South Belfast. The constituency is 2% more Catholic. This will have significant implications for the Westminster and Assembly elections. Taking into account also any demographic changes over 5 years it appears that Alliance could narrowly win a Westminster election. The boundary changes alone would add 2% to the Alliance vote and subtract 2% from the unionist vote and Alliance only lost by 4% in 2019. There could also be a net loss of one unionist seat in an Assembly election.

South Belfast
The wards transferred to East Belfast are mentioned above. The Moneyreagh and Saintfield wards are added from Strangford and Drumbo from Lagan Valley. Overall, there is no demographic change and thus no expected changes in the election results.

West Belfast
Derryaghy South is added from Lagan Valley. Forth River is transferred to North Belfast and the new Shankill and Woodvale wards are added from North Belfast. The new constituency is 5% less Catholic. In an Assembly election this would increase the total unionist vote to 16%, just short of a quota of 16.6%.

These are the initial proposals. Anyone can submit comments and alternative proposals until mid-December. These changes do not appear to be major or controversial in most constituencies. The Boundary Commission must devise constituencies with an electorate between 69,724 and 77,062 voters. Because the Northern Ireland electorate is lower than the UK average per constituency the average per constituency is closer to 69,724.
I expect that the major objections will involve the Strangford and South Down constituencies. The 5 wards of the Downpatrick DEA within the current South Down will be transferred to Strangford. These could be transferred back to South Down and Ballynahinch, Kilmore, Saintfield, and Moneyreagh could be transferred back to Strangford. With any changes, Strangford will need to add one ward from the current South Down constituency (in addition to the part of Crossgar ward currently added to Strangford).
The electoral implications will be much clearer after the Assembly elections in May 2022. Also, the census results will be available in another year. Since the new wards do not align exactly with the old wards the census will provide more accurate demographic details when available.