Well, that's one way to report an opinion poll
Welcome to the 143rd edition of The Week in Polls (TWIP), which takes a look at an …. interesting … Mail on Sunday front page headline about an opinion poll.
Then it is a summary of the latest national voting intention polls and a round-up of party leader ratings, followed by, for paid-for subscribers, 10 insights from the last week’s polling and analysis.
This week, that includes whether people think it is likely Britain rejoins the EU in the next 20 years.
Before we get down to the polling business, my quarterly interlude for an update on what the voters are doing in council by-elections:
Want to know more about political polling? Get my book Polling UnPacked: the history, uses and abuses of political opinion polling.
A simple headline, but was it an accurate headline?
Writing about political polling every week for 143 weeks now (and often enough before that too) has seen me gradually refine a core list of beliefs about political polling.
Top of the list is that polls are more accurate than many people realise, as it is the rare polling misses that get much more attention than the regular polling hits.
Second is that the answer to most problems with political polling is to have more polls, please and more coverage of polls, please.
Third, and this time’s focus, is that problems from questionable polling stories getting into circulation are far more often caused by questionable reporting rather than by questionable polling.
On that note, here is the Mail on Sunday’s front page from last weekend:
So too on the newspaper’s website:
Read on, and you get a clue something may be off with that headline:
Nearly a third of all Britons polled in the 'state of the nation' survey expect the Labour leader to last another year at most.
The data is from a reputable pollster, Deltapoll, and from a recent, properly conducted poll.
But hang on, you may well think. How do you go from nearly a third thinking he will last at most a year to a headline saying the poll predicts he’ll be gone in a year?
As a further graphic reveals and as the pollster’s own data tables later revealed in more detail, it is hard to figure out the answer to that:
Fair enough, you can add together ‘under one year’ and ‘one year’ to get a ‘within a year’ total. That gets us to 31% saying ‘within a year’, with 56% picking a time span more than a year.
You could even exclude the don’t knows (as that is pretty common with poll reporting), which gets the 31% up to 36% saying within a year. But that is up against 64% saying longer than a year when you exclude the don’t knows there too.
So, by 56%-31%, or by 64%-36%, people think Starmer will last more than a year as Prime Minister.
In what way is that a poll that predicts he will be out in a year?
(The headline is so out of kilter with the data we do not need to get into the question of how useful the polling question is as a prediction anyway, i.e. whether it is one that gets genuine predictions from the public or one that gets ‘expressive’ answers, that is where people give the answer that is most inline with their political views regardless of whether or not they literally believe their own answer.)1
Some of the truth may have been further down in the newspaper story, and more of the truth in the pollster’s tables, but the big headline that grabs attention? The big headline which is what gets put prominently on screen in broadcast TV’s newspaper reviews, the big headline that is the one prominent in the newspaper section in shops that people glance at as they pass, and the big headline which is what those grazing social media or the paper’s own website will see? It is very hard to see how reading the headline would give you an accurate version of the story.2
Especially as elsewhere in the poll, the findings are that Labour is ahead on voting intention, and that lead was up on the previous poll by Deltapoll (a seven point lead, up by three).3
Two other things are of note about the contrast between the headline and the full information in the pollster’s data tables.
One is that the headline from the front page started circulating on the Saturday night. The British Polling Council’s rules for polling transparency only require the polling tables to have been published in full by the end of the working day on Tuesday. Deltapoll did better than that, putting the tables up on Monday, the first normal working day after the story broke. Even so, truth on Monday after headline on Saturday evening is rather sluggish.
The other is the fun you can have looking through the full poll, thinking of the other headlines that, with a different editorial line, the newspaper could have put on its poll:
NEW YEAR, NEW CHEER FOR KEIR AS LABOUR’S POLL LEAD SOARS4
CANCEL HIM! MASSIVE PUBLIC BACKING FOR CANCELLING OFFENSIVE CELB5
HANDS OFF HARRY! BRITS TELL TRUMP TO LEAVE OUR PRINCE ALONE6
WE LOVE EU: BRITS SNUB BREXIT AND WANT TO MAKE UP WITH EU7
Back though to the headline that did appear. Some of the follow-up reporting was rather better than the headline, as with Politico’s reporting of the poll:
A new poll for the Mail on Sunday predicts that Starmer will be out of Number 10 within a year, kind of. Some 31 percent of those surveyed say they expect Starmer to last a year or less in the job. The poll still puts Labour seven points ahead of the Tories who sit only one point ahead of Reform.
That is a rather gentle and lowkey correction compared with the shouted certainty of that front page headline.
Moreover, it is now a week since the headline appeared and, at time of writing, it has not (yet?) been changed online. It still reads “Keir Starmer will be out of No 10 within a year, poll predicts”.
I was tempted to add as a final point that there is a general problem with British political media avoiding ‘dog eat dog’ type coverage. When a journalist or outlet does something well deserving of scrutiny, everyone else tends to look away. But Stephen Bush should know more about that than me and clearly has a very different view (and some journalists were rather caustic on their own social media accounts about that front page). Even so, it is fair to highlight that there have been no correcting stories I can find from others of equal - or even half as much - prominence.
To end then as I began, the problem with that polling story wasn’t the poll or the pollster.
Voting intentions and leadership ratings
Here are the latest national general election voting intention polls, sorted by fieldwork dates. Note that once again Find Out Now has both grabbed the headlines and is the outlier, consistently showing a better position for Reform than other pollsters.
Next, a summary of the the leadership ratings, sorted by name of pollster:
For more details, and updates during the week as each new poll comes out, see my regularly updated tables here and follow The Week in Polls on Bluesky.
For the historic figures, including Parliamentary by-election polls, see PollBase.
Catch-up: the previous two editions
My privacy policy and related legal information is available here. Links to purchase books online are usually affiliate links which pay a commission for each sale. Quotes from people’s social media messages sometimes include small edits for punctuation and other clarity. Please note that if you are subscribed to other email lists of mine, unsubscribing from this list will not automatically remove you from the other lists. If you wish to be removed from all lists, simply hit reply and let me know.
Winter Fuel Payment cuts seen as more damaging than Partygate, and other polling news
The following 10 findings from the most recent polls and analysis are for paying subscribers only, but you can sign up for a free trial to read them straight away.
Polling from More in Common tells us that by 51% to 16% people say cuts to
Keep reading with a 7-day free trial
Subscribe to The Week in Polls to keep reading this post and get 7 days of free access to the full post archives.