jump to navigation

Weekend polls April 21, 2019

Posted by WorldbyStorm in Uncategorized.

I’m away this weekend but caught this:

A Sunday Times Behaviour and Attitudes Poll suggests Fianna Fáil has overtaken Fine Gael for the first time since July 2017.

It suggests Fianna Fáil is up four points to 29%, while Fine Gael drops three points to 28%.

Sinn Féin increased by two points to 21% and the Independent Alliance is up two to 4%.

Labour is on 4% (down one point), the Greens are on 1% (down two points) and the Social Democrats are on 1% (up one point).
There was no change for Solidarity-People Before Profit which is on 1% and Independents and Others are on 10% (down two points).

Whereas RedC in the SBP has a different reading entirely:

FG 33 +2

FF 23 -2

IND 16 +1

SF 14 +1

LP 5% NC

GP 3% NC

SD 2% -1


PBP 1% – 1

Aontú makes its first poll showing at 1%.

Make of these what you will.

Paddy Healy points to the contradictions here in comments.


1. Paddy Healy - April 21, 2019

Both Polls Can’t be Right as Margin of Error Claimed is 3% in Both But Difference Between Polls is FF 6, FG 5, Sinn Fein 7 https://wp.me/pKzXa-jh -Red C Final Prediction was also way out in 2014 European Election
B&A Sunday Times Poll, April 21 (%)
FF 29 (+4), FG 28(-3), SF 21(+2) Labour 4(-1), Others 18 Margin of Error 3%
Red C S, Business Post Poll April 21 (%)
FF 23(-2), FG 33(+2), Sinn Féin 14(+1), Labour 5(NC), Others 25 Margin of Error 3%
RED C Was Way Out in Euro Election 2014
Sinnn Féin won 3 seats in the 26-counties heading the poll in Dublin
Euro Dublin Constituency Outcome SF 23.6% Lab 7.4%; Final Red C Prediction SF 15% Lab 13%


WorldbyStorm - April 21, 2019

And the divergence between the two wrt Inds/smaller parties – one has them close to a third cumulatively overall and the other barely 20%. Paddy which do you think is more robust of the two polls (or least wrong?)


2. Paddy Healy - April 21, 2019

Red C is by far the more wrong. Red C , in addition to other matters, doesn’t publish a raw poll or genuine core vote. I will dig up my analyses of the two methodologies and post them here when I get back to Dublin tomorrow. But your question is the most relevant-“who is least wrong?” But because B&A publishes genuine raw poll data one can correct their mistakes-not so with Red C Red C only polls and records the votes ofthose who say they are 80% to 100% likely to vote Red C admits this and claims it is a correct methd!!!! I track the raw poll of B&A each month to see what is happening. Even for good polls, they are very unreliable for small parties such which are not known everywhere and more so for new parties such as Aontú.

Aon’ Wha’?


WorldbyStorm - April 21, 2019

Thanks Paddy – much appreciated as always. A lot of food for thought there


3. Paddy Healy - April 21, 2019

Raw Polls B&A In First 4 Months of 2019
The only change that is greater than the 3.3% Margin of Error is the rise in support for FF of 5% between March and April 2019
April 21 2019 (%)
FF 22 FG 22 SF 16 Lab 4 Others 13 Undecided 24
March 20 2019 (%)
FF 17 FG 24 SF 17 Lab 3 Others 14 undecided 25

Feb 19 (%)
FF18 FG 22 Lab 3 SF 16 Others 14 Undecided 25
January 19 (%)
FF 20 FG 21 Lab 2 SF 18 Others 13 Undecided 25
Note that Undecideds continue at a high of 25%+


4. Paddy Healy - April 22, 2019


I carry at the link the B&A Raw Poll for the poorer half of the population for the first four months of 2019.
Bearing in mind that the margin of Error is 4.6% which is above the figure of 3.3% for a poll of a sample of the full population we note that:
Between March and April the FF share rose from 16% to 23%, well above the margin of error. This is in line with the rise of the Fianna Fáil national poll by 5% in the same period which brought it level with FG. Note that FF had also been at 23% in January.
FG has experienced no change in support in this cohort varying only between 16% and 18%
The Labour Party scores 3%, 3%, 3% and 4%. This is a disastrous score for Labour,bearing in mind that this is a poll of the poorer half of the population. If repeated in the Local and European elections which are only a month away, it would lead to an electoral debacle
Sinn Féin has been the leading party among this cohort in January, February and March 2019. Now in April it has been overtaken by Fianna Fáil. SF’s highest score of 24% occurred in January and its April score is 19%. This is a little more than the margin of error. Its average over the 4 months is approximately 20% ,WHICH IS 6 TIMES The LABOUR PARTY AVERAGE over the same period.


5. Paddy Healy - April 22, 2019

Why caution is required in relation to the conclusions of Red C Polls https://wp.me/pKzXa-jh

RED C Normally Underestimates Sinn Féin Support. This is because its methodology under values political support among the poorer half of the population where Sinn Féin support is particularly strong . This may also explain the gross underestimation of FF support in the April 2019 Red C poll. On the same day Red C had FG 10 points ahead of FF, while B&A had FF one point ahead of FG!! Traditionally, FF has had much stronger support among the lower social categories than FG.
Red C Predictions Were Totally Wrong for European Election 2014, particularly in Dublin.

Sinn Féin won 3 seats in the 26-counties heading the poll in Dublin
Euro Dublin Constituency Outcome SF 23.6% Lab 7.4%; Red C Prediction SF 15% Lab 13%
RED C Final Pre-Election Poll and Prediction 1/05/2014
Richard Colwell “Are Sinn Fein going to win seats in all constituencies? While Sinn Fein poll relatively well in the three constituencies, it is not well enough to guarantee seats in any of the three constituencies. The main reason is that while first preference support for the Sinn Fein candidates is pretty much in line with their national standing, this is not enough to secure seats on its own, and the candidates are not as transfer friendly as other candidates they are competing with”—Red C Website
Sinn Féin headed the poll in Dublin and won seats in the other two constituencies

From Red C Site
” Interviews were conducted across the country and the results weighted to the profile of all
adults. A further past vote weighting is included that takes the recall for how people voted at the last election, compares this to the actual results and weights the data between the two.
Vote intention results are based on those who will actually go and vote, using a 10 point scale, where 1 is not at all likely and 10 is very likely, those rating 8 to 10 are included as being those who will definitely go and vote”

As can be seen above ONLY those who say they are very likely to vote are included EVEN IN THE CORE VOTE. THIS PROCESS ALONE BIASSES THE OUTCOME IN FAVOUR OF THE STATUS QUO favouring COMMITTED PARTY SUPPORTERS and ,arguably the “better off”. RED C does not provide a genuine core vote on its website. On the other hand, B&A provide a core vote on Banda.ie which is ACTUALLY a totally Raw Poll.
Red C, not content with merely eliminating the undecideds from it’s already filtered core vote to get an outcome, but takes that outcome and moves it towards what that party or group got in the last general election TO GET A FINAL OUTCOME.
RED C will of course argue that this methodology has proved a good predictor of election outcomes in the past. In my view, while that may be true in stable political times, it is bound to fail to detect rapid and sharp political changes. The totally wrong RED C prediction of the results of the 2014 European Election,particularly in Dublin, is a prime example.
Unlike B&A, RED C has failed to detect the process over 3 months, March to May 2018, in which Sinn Féin has caught up with and surpassed FF among the population as a whole. Unlike B&A, Red C has also failed to detect the fact that in April Sinn Féin became, for the first time ever, the strongest of ALL PARTIES among the poorer half of the population(Social Strata C2DE). Not surprisingly, RED C has also failed to detect that Sinn Féin has stretched it’s lead over all parties among C2DEs in the May 2018 Poll
Now in 2019, Red C has failed to detect a sharp rise in Fianna Fáil support as discussed above
No mention of these Weightings in Sunday Times Report on B&A Poll 2018
B&A Adjustment (Similar But Not Identical to Red C “Adjustment”)
Positive Weighting for All who state that they would definitely vote
Positive Weighting of all those who definitely stated who they would vote for in the next general election AND who they voted for in the last election, one quarter of the way between voting intention and the outcome of the last general election


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: