Sign in. It’s quick, free and it’s up to you.
An account is an optional way to support the work we do. Find out more.
OVER 17 MILLION British people have voted to take the country out of the EU.
But just yesterday, pollsters said the opposite. A final poll of polls said it was 52-48 in favour of remain.
A day of poll survey of 3,000 voters said the same. Nigel Farage and Boris Johnson conceded they would lose the vote.
But, as with last year’s British general election, some pollsters were wrong. In this case, wrong by a margin of 1.3 million votes.
But, were they really that bad? No.
The reality is that polls had for weeks gone back and forth and had always predicted a close-run race.
As recently as yesterday, Huffington Post’s poll of polls tracker put the sides just 0.5% apart, with 9% of people yet to make their minds up.
Data guru Nate Silver, editor of FiveThirtyEight, said the real issue was why so few people believed the polls:
Polling
One issue that has cropped up is the disparity between phone and internet polling.
Harry Carr, the Head of Sky Data wrote today that there was a major shift in how people said they intended to vote when on the phone and at their computer.
“Online polls had consistently shown a close race throughout the campaign.
“Phone polls had at first shown a far bigger Remain lead, which had closed suddenly in the penultimate week of campaigning amid methodological changes.
“Their final polls had shown the Remain campaign ahead again, but the online polls still showed a close race – with two pollsters, TNS and Opinium, predicting a Leave victory.”
The Telegraph’s expert, Professor John Curtice, had warned that polls were wrong throughout the campaign and said last month’s polling showing the Remain camp in the lead was down to the type of polling.
“In May people were saying Remain were making progress; they weren’t, there were just more phone polls.”
YouGov’s analysis says the two methods can be quite different.
“In person-to-person dialogue people feel pressured by the expectation of an opinion, and then it’s much more likely to be for the status quo rather than change.
We believe this accounts for most of the difference between the methodologies. It doesn’t make one right or wrong. Indeed, judging how many people really have an opinion and will turn out to vote is the big question mark hanging over this campaign, and we expect to find out more and more as we get closer to the moment of decision.
Bookies
In the last week, there had been some argument over whether to believe pollsters or bookies.
In turns out the pollsters were closer to being right. Bookies had pegged a remain win for some time, with Leave chances topping out at around 42%.
Last night, Ladbrokes’ were offering 1/4 in favour of remain and 3/1 in favour of leave.
Surprise
Given that Farage and Johnson appeared to concede, it begs the question what changed?
Leading pollster Ben Page of Ipsos Mori, told RTÉ’s Morning Ireland the result was a “vote of anger”.
Most polls had predicted the result would be extremely close, but it was mainly predicted that the Remain side would win the vote.
Page said he believed people had voted on immigration rather than economic issues.
He noted that the turnout of 72% was higher than expected, and that more working class people had turned out than in previous elections and referendums.
To embed this post, copy the code below on your site
also , I don’t think it should be called ” bettering ” their exams results , it’s questioning their results to make sure they were graded properly .
Suzie, of course people are appealing to better their result. If they thought that the examiner had been generous to them do you really think they’d appeal in order for the exam to be “graded properly”?
sloop ,yes to better their result because the students believe they did better than the results they were given so therefore appeal , but if it had been graded properly in the first place then they wouldn’t have to appeal .
But its not called bettering?
Agreed Suzie, but, not wishing to sound pedantic, the article is about the results of the appeals not why the students had to appeal. Having said that, the fact that there were so many appeals raises serious questions which need to be answered
so nearly a fifth of results appealed were wrong? that’s quite startling
Over 25% of Biology was upgraded. What a poor marking performance. And let’s not forget those marking Biology were the ones teaching it too…. scary.
25% of appealed results were upgraded, appealed results are not an accurate sample of exam papers, it is in fact a very small percentage of total papers.
25% of Biology was upgraded ? – where are you getting that figure from ? 245 out of 25,596 is not 25%……..
Now if you said 25% of those appealed, that would be different………..
well they don’t generally upgrade non appealed papers now do they. Think about it for a while…. 25% of all appealed Biology papers were upgraded. They are a representation of both the standard of marking a teaching.
I understand that S ReiIIy, but I think some thought should be put into reviewing a broader sample of appealed and unappealed papers to investigate what the actual % is, given that the percentage of appealed is so high.
How can so many results be wrong? Will the teacher that gave these wrong results be marking next year?
Course they will. You can’t punish anyone in this country
Except non public or state and semi state workers. we’re fair game.
2000 wrong out of almost 390,000 is a pretty good result. Don’t forget that for many exams what is right and what is wrong can be a matter of opinion.
1,822 upgraded yet only 3 downgraded, I somehow find it hard to believe that so many were marked mistakenly in one direction only.
There’s always going to be more upgrades as you would generally only appeal a result if you think you should have been marked higher after viewing the script yourself, and often with a teacher. If you thought you were lucky to get what you got surely you’d leave well enough alone and not appeal it?
I never knew there was an exam on One Direction Pat? If there was I’m sure a lot of teenagers scored well on it !!!
Lot of good it does for students who may have missed out on a college place due to erroneously marked papers when the term started 6 weeks ago for most colleges. Shocking number of papers marked incorrectly too it seems.
Is 1 in 200 really “shocking”? Bear in mind that many papers have a lot of wiggle room for answers to be “right” or “wrong”. Take Economics for example, one I got checked. For one question on the advantages of Foreign Direct Investment one could say:
Increases Employment
Increases State Revenue through Taxtion
Decreases Social Welfare Spend
Increases Consumer Spending
Decreases State Healthcare costs.
You could get anywhere between 5 and 30 on that answer, depending on the corrector and marking scheme. Some will come along and consider them all separate valid points, another could say that they are all spins on the same answer.
@David Hanlon. I agree that the results have come out too late. However, if you look you will see that the highest number of exams that were marked wrongly apply to Ag Science and that was 1.8%. The total on average is 0.47%. That means that 99.53% of all papers were deemed to have been marked correctly. These figures refer to the overall number of exams corrected and not to the appeals.
6 weeks is far too long to have to wait for the appeal results.
All college places have to be postponed till next year now.
It’s very important for students waiting for appeals to accept a relevant course this year so that it might count for the original course they wanted eg taking science while your hoping for medicine
Them percentages are all wrong
Did my leaving this year and failed business was shocked when I seen this so decided to view my papers , when I seen I got 38% I walked out of the room, it was the most frustrating thing ever since I did 9 months of grinds for 6th year luckily enough I got a course I’m enjoying but 2% off a pass is horrible to see
what about the junior cert appeals when can we get the statistics on those?!
have your say