Express poll highlights Brexit poll pitfalls

by Sam Ashworth-Hayes | 12.02.2016

If you took part in the Express poll that “proves” Britain wants out of the Eurozone, InFacts has good news for you: We’re not in it, don’t have to be in, and are unlikely to be in the near future.


In fact the poll asked about EU membership, and while the failure of the Express to distinguish between the EU as a whole and the currency union is amusing (and now corrected online), the poll’s deficiencies go much further.

The poll doesn’t reflect the views of two large groups —those who don’t read the Express, and those who read it but did not respond to the poll.

All the poll tells us is that some 8,000 Express readers might want Britain to stay in the EU. Or perhaps not even that many – anyone can register their opinion, and some may have done so multiple times.

Want more InFacts?

Click here to get the newsletter

Your first name (required)

Your last name (required)

Your email (required)

Choose which newsletters you want to subscribe to (required)
Daily InFacts NewsletterWeekly InFacts NewsletterBoth the daily and the weekly Newsletter

By clicking 'Sign up to InFacts' I consent to InFacts's privacy policy and being contacted by InFacts. You can unsubscribe at any time by emailing [email protected]

Even the results of surveys by respected pollsters can vary, in part because the sample is small – often just a thousand people. A poll can obtain a different result simply by asking a different group of people the same question.

Online polls on the EU referendum show Remain and Leave to be neck and neck. Telephone polls—which considerably outperformed online polls in the General Election—put Remain ahead by a considerable margin.

The “poll of polls” from WhatUKThinksEU on our site currently shows a small lead for the Remain campaign.

Tomorrow, it may show Leave ahead. But we shouldn’t read too much into it. The poll of polls is an average of the most recent results from a group of pollsters.

This reduces the randomness effect of individual polls but does not allow for each survey’s different methodology, and it is not the same as polling a larger much more representative sample.

Edited by Yojana Sharma

Tags: , Categories: Articles