Adrian Kavanagh, 5th February 2011
The Red-C poll to be published in the 6th February edition of the Sunday Business Post estimates party support as follows: Fianna Fail 17%, Fine Gael 35%, Labour 22%, Green Party 2%, Sinn Fein 13%, Others 11% – based on these poll figures, my constituency level analysis (as described in previous posts) estimates seat levels for the parties as follows: Fianna Fail 25, Fine Gael 71, Labour 39, Green Party 0, Sinn Fein 14, Others 17 (including 6 United Left Alliance candidates – overall 10 “Left” leaning Others/7 “Right” leaning Others)
Seat levels for different coalition options would stand as follows: Fine Gael/Labour 110 seats (majority of 54 seats), Fine Gael/”Right-leaning” Independents-Others 78 seats, Fine Gael/Fianna Fail 96 (majority of 26 seats), Fianna Fail/Labour 64 seats, Fianna Fail/Labour/Sinn Fein 82 seats, “Left Coalition” 68 seats, Fine Gael/Green Party 64 seats, Fine Gael/”Right-leaning” Independents-Others/Green Party 78 seats. As suggested above, the Others grouping is a very broad church, involving a number of left-leaning independents and smaller parties (including the United Left Alliance grouping), as well as business/reform independents and disaffected former members of Fianna Fail, Fine Gael and Labour. The model would guesstimate that 10 of the 17 seats allocated to Others would fall to left-leaning independents/smaller parties, including six seats for United Left Alliance candidates (Barry – Cork NC, Daly – Dublin N, Collins – Dublin SC, Higgins – Dublin W, Boyd Barret – Dun L, Healy – Tipperary S).
(Outside of this, and other polls analyses, my own take on the constituencies and likely results in these may be viewed here.)
The proportion of seats won by parties in Irish general elections does not tend to measure up exactly to their actual share of the first preference vote (although the system is decidedly more proportional than its UK counterpart), mainly because party’s first preference votes need to be filtered through the system of Irish electoral constituencies. In order to address this question, in my constituency-level analysis I attempt to estimate what party first preference votes would be in different constituencies, assuming a uniform swing across all constituencies – a similar (proportional) change to the national swing in party vote shares in all constituencies. How does this work? Well, for instance, Fianna Fail’s share of the vote is estimated in this poll to now stand at just two-fifths of the level of support that the party won in 2007, while Labour support is estimated to be just under two and half times the vote share that the party won in the last election. So for any constituency, the Fianna Fail support levels would be estimated at two-fifths of the percentage support levels achieved by Fianna Fail in 2007 in that constituency and the Labour support level in that would be estimated at just under two and a quarter times the 2007 percentage support levels. Support levels for all other parties are calculated in the same manner, and then these constituency estimates are standardised so that these all add up to 100.0% exactly in all constituencies.
This of course is a very rough model, and cannot take account of the fact that changing support levels between elections tend to vary geographically, even within specific provinces or regions, and also the impact that territory transfers, brought in by the 2007 Constituency Commission report, would have on vote share (although changing numbers of seats are taken account of). The knock on effect is that constituency support estimates for parties in some constituencies may be significantlyover-estimated (e.g. Labour in Kildare South) but over-estimates may be balanced out by under-estimated for other constituencies meaning that the overall aggregation may be closer to what the overall standings for parties would be should the actual election results emerge as exactly the same as these poll figures. Based on these estimated figures, I proceed to guesstimate the destination of seats in the different constituencies in each of the different region/provinces and then to aggregate these to complete the national estimates.
My estimates as to what parties’ shares of the first preference votes in the different constituencies would be:
|Cork North Central||13%||31%||23%||1%||13%||18%|
|Cork North West||26%||59%||13%||2%||0%||0%|
|Cork South Central||20%||40%||22%||4%||11%||3%|
|Cork South West||18%||48%||22%||3%||10%||0%|
|Donegal North East||21%||29%||11%||1%||33%||5%|
|Donegal South West||21%||30%||6%||1%||40%||2%|
|Dublin Mid West||15%||27%||26%||5%||19%||8%|
|Dublin North Central||18%||33%||16%||2%||7%||24%|
|Dublin North East||15%||28%||31%||3%||23%||0%|
|Dublin North West||18%||11%||40%||1%||26%||4%|
|Dublin South Central||11%||15%||38%||2%||16%||18%|
|Dublin South East||13%||26%||39%||6%||9%||8%|
|Dublin South West||14%||22%||37%||1%||19%||7%|
|Kerry North-West Limerick||10%||34%||19%||1%||31%||4%|
Based on these constituency estimates, I would guesstimate seat levels to fall as follows:
|Cork North Central||2||1||1|
|Cork North West||1||2|
|Cork South Central||1||3||1|
|Cork South West||2||1|
|Donegal North East||1||1||1|
|Donegal South West||1||1||1|
|Dublin Mid West||1||1||1||1|
|Dublin North Central||1||1||1|
|Dublin North East||1||1||1|
|Dublin North West||2||1|
|Dublin South Central||1||2||1||1|
|Dublin South East||2||2|
|Dublin South West||1||2||1|
|Kerry North-West Limerick||2||1|
By and large, this guesstimation would not take account of candidate selection factors – i.e. number of candidates being selected by the different parties, with the exception of Labour (Kildare South), Sinn Fein (Donegal South West) where the model could allocate these parties two seats but for the fact that these parties are running just one candidate. There are a number of constituencies where the model is allocating one seat or two seats to Fianna Fail, but the splitting of this vote between multiple candidates could put these projected seats very much at risk, especially if there is vote leakage involving the intra-party transfers. On the other hand, by running fewer candidates than in 2007 in most constituencies, the likelihood would be that the party vote would be lower than these projected figures given that running one fewer candidate means missing out on the local/personal votes that candidate would have won for the party. It’s Sophie’s Choice for Fianna Fail – run fewer candidates and depress the party vote further so that even the smaller candidate pool is missing out on the constituency seats, or run more (or rather the same number of) candidates and run the risk of losing out on seats that might have fallen to the party based on that vote share? In different constituencies Fianna Fail have opted for different strategies. In some constituencies, akin to the Russian army facing the rampant Napoleon in 1812, they have gone with the first option and went for a salted earth or burnt earth strategy, where the party has surrendered one of the seats they won in 2007 in the hope that they can take something from the election. Examples of this include Cavan-Monaghan, Cork North-Central, Donegal North East, Dublin South and Limerick. In other constituencies the alternate approach has been chosen; Fianna Fail candidates have opted for a high noon strategy – in which neither of two incumbents take the plunge to go quietly into the dark night and both decide to go out fighting (e.g. Dun Laoghaire) – or a man-marking strategy – in which the party’s selection strategy is effectively mirroring that of Fine Gael (e.g. Longford-Westmeath). Most political scientists and analysts would opt for the strategy of running fewer candidates and edge for the burnt earth approach. But as an electoral geographer I must offer a counter-view and hazard the warning that running fewer candidates means missing out on the thousands of party votes that an added, sweeper, candidate could bring in – as shown by the example of Galway East in 2007 – furthermore surrendering seats means surrendering further territory to a rampant opposition who may feel further emboldened to go for the jugular.
30 thoughts on “Red-C/Sunday Business Post poll, 6th February 2011 – seat estimates and Fianna Fail selections”
We’re looking straight at a rainbow government to include Fianna Fail with Eamon Gilmore as Taoiseach and Micheal Martin as Tanaiste/Minister for Finance.
Obviously projections will differ depending on the methods used, but does it not worry you slightly that your method sees the LP seat-to-FPV ratio reduce by about 10% when more than doubling their vote, despite larger parties tending to have a better ratio for this (and the greater number of potential LP transfers)? The FG ratio by contrast appears to increase by 9%, despite the disappearance of the PDs as a source of transfers (and the muscling in by LP, whom previously were eliminated before the last count in most constituencies and transferred largely to them also).
Also, the figures in your other projections seem to be out of synch compared to regional figures (esp LP in C/U).
Obviously I know how difficult it all is, given I do it myself, and one never knows whose method will work best *this time* until it actually tested against an election, but do you think these headline deviations might indicate the seat totals might ‘drift’ from the projections you have here?
I admire anybody who publishes his forecasts at this stage of the campaign, but the polls are remarkably congruent. Like most people I find it hard to believe in the FF implosion, but this evidence is quite persuasive. I really can’t see any grounds for Michael McGrath’s prediction. Why on earth would anybody want to form a government with FF?
Because the Office of Taoiseach is so powerful that no TD can resist it , men would sell their souls to the devil for it, so you can’t blame Eamon Gilmore – he only has to sell his to Fianna Fail. NO PROBLEM!
It would also be powerful for the Irish Labour Party after 99 years in the wilderness, so you can’t blame them either.
In human terms it makes sense. In my own view it would make for the best government we could possibly have, essential in our national crisis.
Also bear in mind that the trade unions would insist in Labour grabbing the Taoiseach power position.
Happy now, EOF ? And btw three weeks is a virtual eternity in an election – even now the differences bewteen FF and FG are looking minimal as FG would do the very same thing , but with Enda Kenny doing it.
Personally I am in the Anybody-But-Kenny camp. so I gather are most people out there. Because the man is the most petulant fellow in Ireland, as he has proved once again in the TV interviews saga ongoing while the economy collapses. It’s all about what he wants, it’s all about what insults he endures eternally , it’s all about Poor Enda.
And . no, we didn’t sell Ireland out to the IMF/EU/Bankers , we didn’t sell the country out , that’s looking at Ireland in terms of economics and nothing else.
But let anybody come and try to take Ireland , and they’ll meet the Fighting Irish.
For instance the environment is more important than the economic sphere. National re-unification is more important too.
We can recover the economics, we can recover the money, but we can never recover the soul of Ireland!
Given that Governments with large initial majorities (FF, Jack Lynch 1977, and FF/Greens/Independents, Bertie Ahern 2007) tend to be unstable, I can’t see a Government with an absolutely huge majority (FG/Lab 2011) surviving for long. Labour would implode/split under the pressures of implementing ECB austerity plans and opinion polls would quickly tell their back benchers that they are about to lose their seats and had better revolt quickly. A FF/SF opposition would be transformed into a Government in waiting within 18 months.
So Michael McGrath’s (tongue in cheek?) suggestion of a Labour/FF/SF/some independents Government is not as surreal as it may seem. I see two problems: 1. FF is actually campaigning to go into opposition and would be egarded as a toxic coalition partner even with Michael Martin. 2. FF would never agree to play second fiddle to Labour as this would destroy their chances of becoming the largest party again after the next but one election.
On this occasion, its not who gets into Government who wins, but in the election after this one (which could be within 18 Months as all possible coalition options arising from the currently predicted vote distributions would be quite unstable. I see FF playing the longer game and profiting from a Labour implosion/split when the next Government folds.
True, Frank, but Realpolitik.
Bertie Ahern admitted to me , in a private chat after I had photographed him on his book signing in The Bookshop , High Street, Kilkenny ( BTW one of the better bookshops in Ireland) that he totally missed the power of being Taoiseach, ” There’s nothing quite like it” , Bertie sighed.
” Poetic Justice , Bertie ” , another photographer suggested –
” And now they want to jail me on top of it all! ” , Bertie flashed back.
But Bertie did say to me and the other photographer present that he always much preferred the company of photographers to “other” journalists, ” they’re out-and-out savages” he continued …
” Say no more, all is forgiven, Bertie”, I flashed back.
Adrian, compare these seat estimates to the results of the local polls carried out for today’s independent. The data are slightly dodgy and because they’re based on phoning landlines mean that they probably have less coverage of the young and the poor – so Sinn Féin and ULA may be underestimated. But despite these they show that Brian Lenihan is probably safe and that FF has a good shot at taking two seats in Cork SC. Going by these figures, a FF seat result in the early to mid-thirties seems more likely.
Yes, Eoin, and two FF , McGuinness and Aylward, to be re-elected in Carlow-Kilkenny, the polls are beginning to make sense and add up now. You could have a straight Labour – FF government yet with Eamon Gilmore as Taoiseach.
Is there any purpose in taking a Sunday Independent/Quantum Research poll as indicative of anything but the Sindo editorial line?
Has anyone manage to find Quantum Research, that most mysterious of polling agencies, yet?
Can anyone even find the exact numbers of the Quantum research online anywhere? The Sindo “analysis” is v. partisan and leaves out a lot of key numbers – and even manages to lame wrong conclusions based on the numbers it does publish. I am compiling a spreadsheet model of local polls first preferences and trying to account for transfer patters etc. – but withour access to the raw data there isn’t much point. What the polls do highlight, however, is that when you present voters with a choice of candidates rather than parties, you come up with quite different results.
I suspect the Sindo is running these polls to help FG with its vote management strategy, as the larger your vote the more importance vote management becomes.
“Quantum Research” don’t seem to have a website. Nor has anyone yet found any other contact details for them. They don’t make data available except for that published in the Sunday Independent. Nor do they seem to do any work for anyone other than the Sunday Independent.
I’d say that they aren’t worth the paper their printed on, except for the fact that they appear in the Sindo.
Of course, Eoin , you’ve hit on one weakness of the polls, a lot of young voters are in flatland with only mobiles, and they would be country folk in jobs in Dublin who could well be satisfied enough in their lives to vote FF like their parents always did!
And I wonder too, Eoin, many of the immigrants with votes, like the large Polish community who have been here since 2004 , and even before that , would have votes – FF always played up to them!( and they wouldn’t have any concern about bankers, sell-outs, bale-outs etc) – would also be in flatland throughout the country, as they are here in Kilkenny , they wouldn’t mostly have landlines either:-). John McGuinness TD even put forward a young Polish woman here for the 2009 local elections, he has the immigrant vote wrapped up so much that he was even made Honorary Chief of an African tribe 🙂
Only resident Irish and UK nationals have votes in the general election.
Remember what happened to Labour’s Spring tide after they put FF back into power after campaigning for reform and a new government (without FF) in the early 90’s?…..Labour got destroyed and punished severely in the next election for that betrayal and Gilmore and co. would do well to remember that harsh lesson. For the good of Irish democracy FF needs an absolute minimum of 5 years in opposition. 10 years would be better as then we could properly cleanse the country of their cancerous influence.
Of course if FF do better than the national poll projections then the size of the FG=Lab supermajority starts to come down towards more manageable levels. There are at least four varieties of independent ( left alliance; right-wing rebels; FF gene-pool and opportunist deserters and of course single-issue candidates). If SF does as well as the polls predict and attracts transfers from the independents then this could also eat into the FG-Lab majority and possibly ensure better discipline. Historically very large majorities don’t necessarily collapse. Cosgrave managed to get be through most of the twenties without a real opposition. The 1977-81 government was fatal for Lynch’s leadership but FF stayed in power and was really finally undermined by Haughey’s incompetence on the economy, a recurring problem and a direct consequence of populism.
With the diference, Noel, that it would be Fianna Fail putting Labour in power this time round:-)
The use of the quantum research local polls by Sunday Independent is quite outrageous and the conclusions drawn are totally unjustified. The selection of constituencies used is extremely unrepresentative of the country and heavily weighted towards Fine Gael voters. The two Dublin constituencies have unusually high “middle class” populations. Cork South covers the wealthiest areas of Cork City etc. The use of landlines only renders the sample anything but random. The size of sample(400) gives a high probable error (5%). We recall that Independent Newspapers came out in favour of Fiann Fail in the last general election. As pointed out by the Greens to-day, Fine Gael is being bank-rolled by anonymous vested interests. Is the biggest newspaper group in the state now attempting to create a “bandwagon” for Fine Gael in an attempt to protect the wealthy as Fianna Fail did in the current Dail?
Stands to reason , Paddy , they must always have their servants in power.
And the people will be right fools – and the victims of it all – if they vote to let it happen.
But a Labour-Fianna Fail-Sinn Fein-ULA – Independents Rainbow could turn the tables on the Indo-Fine Gael Bankers Set Up Fine Gael yet with Eamon Gilmore as Taoiseach.
It is obvious that the smell of this stench emanating from the pages of the Indo reached Gilmore’s nostrils last week.
Obvious too why Kenny and FG want to get known troublemakers that can’t be depended on like Vincent Browne out of the equation, they stink to high heaven.
Anyone giving any credence to Quantum Research (which doesn’t physically exhist or do any other market research) must be very foolish, and the journalists reporting on them, lose any credability.
The sample size is only part of the story…
What about sample selection, in a proper poll calls/interviews are conducted based on population spread across DED’s within the constituency?
Or weighting to known voter profile, gender, age, class after the event to ensure it is representative?
How about weighting by past voting to ensure the sample is accurate?
Also what about the way the questions are asked.
Too many questions and not enough above board answers screams they are dodgy. Also if they were OK, why spend a lot of money getting a proper polling company to do a national poll at the start of the campaign!
I agree that the company doesn’t exist except in the mind of the sunday independent’s editor and that we know nothing of the non-response or any other biases. But these polls have the advantage of telling us something about individual constituencies and because candidates’ names are given this is an exercise closer to the choice voters actually face in the election. Red-C’s question relates to party and not candidates and so may systematically underestimate the FF vote. They compensate for this somewhat with the reallocation of undecideds on the basis of previous vote, which pushes FF closer to Labour in levels of support. But this is not the headline figure given for FF in Red C polls and the reallocation is given because there are expected to be ‘shy FFers’ not because the question asked is about parties and not candidates.
Also Red C does have some other systematic biases. So by using random digit dialing on both mobile and land line phones, those people with both landlines and mobiles are twice as likely to be selected than those with just one type of phone.
Once we know about the problems with these polls we can make some use of them, so where lots of candidates are close together they might not be useful (Kerry South), but I’m pretty certain that the Dublin West figures giving Brian Lenihan 24 percent probably means we can say is seat is safe (not that many really thought it wasn’t anyway).
I take your point, but the small bias of more chance of people with both landline and mobile to be rung is tiny compared to possible bias in polls which are not in any way controlled. Also professionla pollsters do eevrything they can to eliminate bias. The Red C polls control phone ownership to latest comreg data, so their final polls ends up with the right proportions who have a mobile only, both landline and mobile and landline only.
For all we know the Quantum polls could all be done with people with the surname “Murphy” in one constituency becasue it was eaiser to draw the sample!
There is no indication of any controls over where people are called in the consituency -so in Louth they could all be in Dundalk.
There is also no indication of any control on demographics to match the population in the constituency, so they could all be men.
I agree that constituency polls are useful as they look at candidate effect, but only if they are conducted properly in the first place!
I know assessing these polls and speculating on the election results provides opportunity for great sport and craic, but, and apologies for appearing to seek to dampen the enthusiasm, in my view, the outcome of the election won’t matter one jot. The ‘permanent government’ will remain in place and the game is now being played in the corridors of power in Brussels, Frankfurt, Paris and Berlin.
We do not seem to realise the extent to which our politcians and institutions (both public and private) have lost the trust of our EU partners and international institutions – and of the external investors on whom we will rely, eventually, to provide additional financing of economic activity.
Chancellor Merkel is no longer willing to take anything on trust from her EU partners any more. The first step is the ‘competitiveness agenda’ discussed at last Friday’s European Council meeting. This is simply a front for centralised EZ fiscal governance in the German mode. It is the simple quid pro quo for any expansion of the EFSF to provide some debt relief to the beleagured peripherals.
Whoever is elected Taoiseach will be the EU’s pro-consul in Dublin. It is both tragic and cruelly ironic, when so much attention is being paid to it, that Ireland has lost the sovereignty to pursue and benefit from political and instutional reform.
Richard is absolutely correct about the so-called Quantas Polls
Such spurious polls can influence a time-limited and irreversible election process by creating a “bandwagon effect” for the favored and by reducing the credibility of others
This has implications for democracy
Broadcasters are required to maintain balance
Is there any obstacle to or redress for publishing spurious polls?
The biggest problem we will face is future borrowing if we do not stop all the public statements of burning bondholders!
The problem is that virtually all the opposition parties, especially their leading politicians , are calling publicly every day, even several times a day, for to burn bondholders of some description or other.
You burn bondholders by never talking about it in advance , let alone threatening it publicly every day of the week.
And you stop, breathe in deeply and calculate the reverberations around Europe and the world.
You do not do it at all unless you have assurances from the EU/ECB and the IMF in advance. You do it on a case-by-case determination with the nod from them in each case.
The first thing the new Taoiseach will have to do is give an assurance to the financial world that all the threats of burning bondholders was simply party election manoeuvring, gimmickry and nothing else,just election talk , no more than that.
Then you proceed quitely with the EU and IMF along with you in the determination of every case.
And you also require legal advise concerning the possible attitudes of our Courts if we are sued, that only dragging out the agony and our credit-rating even further down into the mire.
What if you burned a bondholder and he immediately requuired the principle sum back in retaliation? That’s just one very serious question.
IMHO we should keep our honest name as a reliable country at all cost, just as we value our own names.
There is a company called “Quantum Research” registed with the CRO (#430052) and with an address at the Arthur Cox building. If you’re sufficiently curious then you can purchase annual returns from the CRO website.
There is also a company called “Quantum Research Group” with an address at 30 Herbert Street, Dublin 2.
The Irish Times could be interested in bogus Indo polls?
Well said. In my view, what is creating so much volatility in the opinion polls is the extent to which voters are being influenced by the possibility of threatening to press the ‘nuclear button’ on the Euro unless we get get major concessions on the bank debt. All the factions are being pulled into a race to demonstrate how much they will stick it to the EU and the ECB to get a result for Ireland.
This is a dangerous and self-defeating game.
Pingback: Dublin North Central predictions - Page 14