Will a majority of voters in Britain’s upcoming referendum elect to remain in the European Union? (update)

The GJ consensus forecast has started to climb, largely on reports of recent polling data[1]. Color me *a little* skeptical though. The average of the 12 (12!) polls taken in April (as reported by The Financial Times[2]) shows less than a 2% gap between thems that wants to stay and thems that don’t: 42.75% = “stay” and 40.83% = “go.”

EU_referendum_poll_of_polls

Of course that leaves 14.59% who haven’t made up their minds. If you assume that the undecided won’t break in large numbers for either of “stay” or “go” (and I don’t know of any evidence that counts strongly against this assumption), then “stay” gets 51.1% of the vote, and “go” gets 48.8%, with a standard deviation of a little more than 4%. That gives us about a 61% chance of “stay” winning, and that seems too close for comfort.

But maybe the numbers shouldn’t be taken at face value. So I tried weighting the polls for temporal proximity. I gave those polls done within a week a weight of 1.0, those within 2 weeks a weight of 0.75, those within 2 weeks a weight of 0.50, and those done at the beginning of the month a weight of 0.25. The average outcomes barely changed: “Stay” = 51.5%, “leave” = 48.5%, but the lower standard deviation (2.6%) predicts a much higher probability of a victory for “stay”: 72%. And I also tried giving more weight to the polls that have larger groups of respondents, roughly giving a poll that has 3 times as many respondents 3 times as much weight. Yes, I know that’s a pretty naive approach, but what ya want for free? The results were “stay = 51%, “go” = 49%, with a standard deviation of 2.2%, and a 67% chance of victory for “stay.” Things look better for “stay” than they did a month ago, but not as much as the GJ consensus forecast would suggest.

Of course, it might be that the consensus forecast is higher because people expect to see the Renwick effect[3]. I do too, at least to some extent. But my understanding of the Renwick effect involves the shift being visible in the polls in the time leading up to the referendum. Maybe we’re starting to see it, maybe not. I’m moving my forecast up 2 very cautious ticks to “stay” = 73%.

[1] https://www.gjopen.com/questions/149-will-a-majority-of-voters-in-britain-s-upcoming-referendum-elect-to-remain-in-the-european-union

[2] https://ig.ft.com/sites/brexit-polling/

[3] https://www.gjopen.com/comments/comments/194694

Advertisements

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s