ComRes for Centrum Campaign/ Electoral Calculus - 1-7 May 2019

London4Europe Committee Member Michael Romberg works out what the opinion poll findings for the European Elections in London mean for seats. Remember: it’s just a poll – a snapshot that comes with a significant margin of error. The margin is bigger than usual because London respondents are just a small subset of the overall sample.

Data Source:  ComRes for Centrum Campaign/ Electoral Calculus - fieldwork 1-7 May 2019  -  364 replies in London, weighted by likelihood to vote, excluding don't know/ won't say/ won't vote (12% of unweighted London sample).

In addition to the conventional opinion poll results shown in the first row of Table 1, ComRes have performed a Multi-Level Regression and Post-Stratification analysis. That means using the survey to attribute voting behaviour to particular demographics (age, employment status &c); and then applying that knowledge of voting behaviour to the known demographics of the constituency.  You can read more about MRP here.  Table 3 uses the ComRes MRP figures for London. On the basis of that analysis ComRes conclude that in London the Liberal Democrats are the leading Remain party. For consistency with the other polls, in the next tracker I will use the conventional opinion poll numbers in Table 1.

TABLE 1: This table takes the opinion poll findings and turns them into seats using the D’Hondt method (read more about that here). Seats won in each round are shown in bold. Poll numbers are then adjusted for subsequent rounds of seat allocation.

Notes

Brexit = Farage’s Brexit Party

Change UK, formerly The Independent Group of MPs

LD = Liberal Democrats

Note

CAVEATS

• It’s a poll. A snapshot in time. People’s views change. If “don’t know/ won’t say” is large, then their views could swamp small differences between parties once they decide.
• European elections normally have low turnout. Differential turnout amongst supporters of different parties could affect the result compared with a poll. Different polling companies have different methodologies for adjusting for turnout.
• The definition of “London” in the poll may not match the London constituency for the election.
• Polls come with a margin of error. On the highest level figures asking a question of the whole sample a rule of thumb is that polls come with a margin of +/- 3 percentage points (so 45% think this might be anything in the range 42%-48%). London figures are normally a small subset of the poll so the margin of error is larger. A good overall sample size for a poll is 2,000 or so.
• My calculations are on rounded numbers and that might introduce an error when results are close.

TABLE 2: this table works out a counterfactual allocation of seats if the Remain parties had joined on a single platform. The calculation assumes that the vote for the platform would be the sum of the votes for the individual parties. That would not be the case in real life – there are arguments for saying that a platform might do better or that it would do worse than the sum of the parties.

In the case of this poll a common Remain Platform would have increased by one the total seats won by the various parties/ platform.

Remain = sum of poll results for Change UK, Green and Liberal Democrats

TABLE 3:  This table takes the findings from the MRP analysis and turns them into seats using the D’Hondt method. Seats won in each round are shown in bold. Poll numbers are then adjusted for subsequent rounds of seat allocation.

Note: The published rounded figures do not say whether the 8th seat should go to the Brexit Party or to Change UK. I have followed the seat allocation used by Gina Miller's tactical voting advice site Remain United on the assumption that they have access to the underlying unrounded figures.