Subtlety gauge miscalibrated again
A recurring source of amusement for the independence camp is the weekly reader poll in Scotland On Sunday. Time and again the surveys fall victim to deeply-implausible sudden surges in backing for the Unionist option, often in the middle of the night and usually after Yes supporters have drawn attention to less favourable standings.
(The paper’s deputy editor Kenny Farquharson once memorably tried to explain away 25,000 overnight votes – in a poll which had attracted about a tenth that many* in the entire preceding week – as having come from American and Canadian readers, all having inexplicably decided to vote at once on the same day.)
A fairly typical example of the phenomenon, from back in April, can be seen here, but the No campaign’s IT black-ops department appears to have suffered from a bit of an itchy trigger finger this morning and pushed the bounds of credibility a little too far.
32,365 votes for one option? Between midnight and 11 o’clock on a Sunday morning? That’s a real nice job, BT “grassroots” boys and girls. Very convincing. Take the rest of the day off and get yourselves a doughnut.
We actually took that first screenshot at 10.49am, but delayed posting this article in order to give the preceding one a bit of time on the front page. When we looked back at 11.25am, another 2,041 people had cast No votes, compared to just 12 for Yes.
We hit Refresh again at 11.47am and further 1,235 had flooded in for No, against just 14 more for Yes, but when we went to grab another screenshot a few minutes later the poll had mysteriously vanished from the story it had previously been attached to.
It still exists on the SoS servers at the time of writing, and can be accessed directly here, but the only normal page we can now find it on is this one. The voting rate has dramatically slowed, presumably as a direct result – the 58 minutes from 10.49am to 11.47am generated 3,302 votes, the subsequent 40 minutes just 192.
(That supports a hypothesis of some sort of automated spam-voting, which would be foiled at least in the short term by a change of URL for the poll. Which in turn suggests that SoS might suspect some chicanery, yet have chosen not to reset the numbers.)
We can only assume these crudely transparent attempts at social-media manipulation (a pattern study of the “Like” statistics on the ‘Better Together’ Facebook page is also a fascinating exercise, currently being undertaken on an ongoing basis by diligent monitors) is a key tactic of Project Fear. It’s going well.
.
* It’s not easy to Google previous SoS polls for reference, because they’re embedded rather than having specific pages of their own, and many of them seem to have been wiped and reset. But we’ve located a few – the one from two weeks ago has attracted a total of 3,594 votes to date, the preceding one 3,176. One from May on Gordon Brown managed 4,430, and another the same month secured 3,104.
Stupidity is better together
Kenny Farquharson – hehe 🙂
That’s brilliant. However, a source* tells me that it’s not a spambot – it’s just Blair McDougall and George Foulkes sitting at their computers.
*I’ve made this up.
@DougDaniel –
*I’ve made this up.
you sound like a natural for Scotland on Sunday journalism!
Not bad for a paper with a circulation of 38,493. One of course can be charitable and argue that the vast majority of it’s online readers already lean to the no side of things.
Of course it only matters if “Wanting to back the winning horse” is a serious consideration for people in their voting booth. (I’m not saying it isn’t)
You would take it that SoS and the Scotsman are now getting near down to their hardcore Unionist readers but those figures are just silly.
Maybe they were valid postal votes?
The interesting thing about the Scotsman right now is the massive disconnect between its online readers and its pro-union content. The vast majority of comments are very pro-independence, which marks a shift as far as I am concerned.
We can laugh at the poll, but it shows you the lengths BT will go to in manipulating these sorts of things for political gain. I am surprised this poll hasn’t become the main news story on BBC Scotland…..;)
Just more of the same blue/red Tories singing from the same hymn sheet.
There is only one poll that counts. Boy wont they get a shock in September next year when the ‘fat lady’ sings.
Neither the Scotsman or the Herald are now classed as national papers, but if you take the most recent figures on paper sales, with only the Times and i having increased their sales and the rest taking significant decreases in sales, I think it is pretty safe to extrapolate that the Scotsman editor must be shitting himself right now.
Desperate measures and all that. It will be interesting to see the next regional sales report. Twice yearly I think?
It pushes the bounds of credibility that the NO Campaign actually have 32365 supporters!!
Daft nationalists! Its easy to explain. The sum total of the newspaper readership are assumed to have voted no, even though some may not have a computer, some didn’t bother and some are dead.
The figures are now————–Yes–2,559——6%
————–No–38255——94%
“The figures are now————–Yes–2,559——6%
————–No–38255——94%”
Now 10%-90% (4,400 – 40,000), suggesting that if you strip out the 38,000 obviously-dodgy votes, Yes is in fact leading by around 2:1.
It’s my birthday today – so according to SoS they must have all come at once?
Why is their title of today’s ‘poll’ written as “24/06/2013” – Jam Tomorrow I presume.
Clarinda
Perhaps it was meant to appear tomorrow.
Happy Birthday
I think Calum hit the nail on the head, The SoS quite rightly assumes that all of its readership is Unionist Heidbangers and most of them are Computer Illiterates. So they quite rightly apply the Curtice formula to weight for this fact. Now we know what happenned to all those missing Postal Votes from Aberdeen Donside as well, no very bright are the Brit Nats!
Love the irony in ‘spam-bots’ fiddling polls on behalf of ‘spam-heads’. Do you not think replicating the sort of support a North Korean dictator would command, might have been a little bit foolish BT?
wish we could put a spoke in their wheel!
From Sunday Times
THE BBC is facing fresh accusations of bias in its coverage of the Scottish independence debate after Craig Murray, a past rector of the University of Dundee, claimed there was repeated use of “negative language”.
Murray, the former ambassador to Uzbekistan, said a trawl of the BBC website revealed “repetitive coupling” of the words “independence” and “warning”. In contrast, he said, there was an absence of headlines that promote positive claims about Scottish independence.
Murray, who backs independence, also claims the BBC’s flagship shows Reporting Scotland and Newsnight Scotland have “never, never been led by a positive story about independence. It has been led on dozens of occasions by the negative”.
BBC Scotland has denied Murray’s claims. A spokeswoman said: “We reject any suggestion of bias.” Nevertheless, Murray has requested a meeting with Chris Patten, chairman of the BBC Trust, to discuss the alleged “anti-Scots propaganda” which “is moving beyond the risible towards the truly chilling”.
“It astonishes me that even the use of the most blatant state propaganda techniques by the BBC do not result in any serious reaction from the political establishment,” said Murray.
He has also urged Alex Salmond, the first minister, to request the intervention of the Office for Democratic Institutions and Human Rights, an arm of the Organisation for Security and Co-operation in Europe, to monitor the referendum and media coverage in the run-up to September 2014.
Last year, Salmond raised concerns with Patten over the BBC’s use of words such as “separation” and “separatism” instead of independence, on the grounds they implied an independent Scotland would be left isolated. Shortly after, it was claimed the BBC issued a guide telling staff not to use these words.
A spokesman for Better Together, the pro-union campaign, said: “To suggest that the BBC is biased is nonsense.”
Was this and the other polls, an attempt to undermine the credibility of the democratic process?
Ah but that sudden upsurge probably due to exposure on here? maybe best to abstain, the more silly it continues to look the better?
Do you think if we just left it for a day or two the number of No`s would rise to 5.3 million? I think that would mean we separatists would all have to pack our bags and leave Scotland to the little Britishers.
Laugh and point and point and laugh (think Donald Sutherland`s face at the end of the 1978 remake of Invasion of the body snatchers).
Since GCHQ seems to monitor the internet in the UK like PRISM at the NSA in the US. They must know who we all are. One suspects they also control things too. Perhaps only online polls and DNS attacks.
It must have been the Truth Team (Alternative Reality Division) voting!
Iain More says:
23 June, 2013 at 2:30 pm
“Now we know what happenned to all those missing Postal Votes from Aberdeen Donside as well, no very bright are the Brit Nats!”
I missed this, what happened to the missing postal votes?
SCED300
They were posted to The Scotsman for their polls?
It’s almost certainly someone just using a simple script. I could write one up in PHP easily and run it to change whatever votes I like (just call the same URL over and over again in a loop without sending the cookie), but I’d be risking my internet connection if they whine to my ISP.
I’ll write something popular, then very unpopular. I hate The Scotsman, it’s a trash paper (okay, two popular things), but I don’t believe they are fiddling their polls. I just think it’s a unionist with a bit of simple scripting knowledge.
over 44000 no’s now.
When this terminally ill newspaper started introducing these particular polls a couple of years back (and having a bit of a geeky IT background) I ascertained that the poll URL could be manipulated through a bit of wget scripting to hike up one of the vote options.
After a bit of fun I gave it up as a bad joke – a laptop sending out N votes per minute is fine but its needed for Tesco shopping as well you know ;-0
Here’s today exposed URL, your average hacker would have a field day with it. A wee bit of scripting to deal with cookies, and watch that vote soar…
http://www.scotsman.com/news/scotland-on-sunday-poll-24-06-2013-7-44698?optionValueBeforeVoting=45299&pollId=poll_7_44698&vote=true&pollIdpoll_7_44698=0&cookieSet=true&pollIdpoll_7_44698QuestionId0=1
link to auldacquaintance.wordpress.com
(During one point in the film Pilger interviews the BBC’s Head of News gathering, and forces her into the following revelation ” What the BBC have the duty to do, is to Report what Government and their representatives are saying”)
Lies, damned lies and insidious Better Together lies.
Well, having now had a look at that SoS article and stopped laughing at it, what can I say?
If you wanted to present a Better Together Unionist fearbomb scare story to anyone who didn’t know what one was just let them see that, it’s a perfect example. It’s got all the ingredients: a hardcore Unionist quoted; lots of ‘scary’ words like ‘warning’ and ‘danger’, talk of terrorists and cyber-warfare, no facts to back up claims, exaggeration of fears such as job losses and a totally negative picture of an independent Scotland promoted throughout the article.
Poor old British Empire. Nothing left but tales of past glories and lies.
As for the poll, I think it may now have dropped down to ‘only’ 85% No… LOL!
I don’t believe 224 YES supporters even read the paper, surely we would have registered 0?
@ Douglas Young
I know Rev Stu does not promote a boycott, but I personally stopped clicking unto their website a long time ago. I would never lower myself to vote in their stupid poll in a million years. They could get 100% No and just look stupid.
“I just think it’s a unionist with a bit of simple scripting knowledge.”
OR mibbies its wan oh they fabled cybernats we keep heerin aboot/
just tae mak an arse oh them?
“They were posted to The Scotsman for their polls?”
fit like? hiv ye nivver heard o a single transferable vote,
fir christ sake man?
The Scotsman as a paper has zero credibility. Its polls and forums are pathetically hackable. It seems completely unable to do anything about the Chinese spam that was filling every thread the last time I looked. Mind you the spam made more sense than the wing nuts that troll the boards. For a paper that used to be a quality broadsheet it now seems to attract swivel eyed loons that are too right wing for the Mail.
I’ve been collecting the stats for Yes and BT main webpages Facebook ‘likes’ since 21/5/13 and, when I noted on the Herald comments that I was, a certain MM became very interested in my ‘analysis’ (I said I gave no analysis other than that the BT ‘likes’ were essentially flatlining).
That night the FB ‘likes’ for BT went up by 904 counts after 10 days averaging 1.7 ‘likes/day. Coincidence? Hmmm?
It would make life much easier if anyone could write a script which could grab those figures automatically. Let me know if you can but I’ll continue to collect manually (trying for morning, lunchtime and evenings).
Robert, I’m not expert on facebook (I dont even use it). But I had a look at this.
If you click on the LIKES box on the front page it takes you to a graph of the number of likes over time.
If you hover your mouse over the graph line it shows you the number of Likes added in a rolling 7 day period (so its easy enough to work out the number of new Likes each day).
This data goes back about a month or so.
The YES likes have been increasing by between 1300 and 2300 per week, every week.
The BT number has been at less than 200 per 7 day period for most of the last month (sometimes as low as a total of 50 new Likes over a 7 day period) EXCEPT for curious jumps of about 500 on 29th May and about 1000 on 13th June. (The graph jumps up to the new higher level for the next 7 days until that increase drops out of the rolling 7 day total).
It could be that this coincides with some ‘campaign’ by BT to persuade people to Like the page – who knows.
I dont know if it is possible to get data on who these Likes are ?
Here are the links to those pages
link to facebook.com
link to facebook.com
“It could be that this coincides with some ‘campaign’ by BT to persuade people to Like the page – who knows.”
This could actually easily be the explanation, every now and then I get a sponsored/suggested link asking me to do just that. I assume that these appear on a great many peoples pages at the same time and it’s not unreasonable to assume that a spike in likes occurs shortly afterwards perfectly legitimately.
I wouldn’t mind seeing some statistics on how many people have been banned from posting there though!
“I wouldn’t mind seeing some statistics on how many people have been banned from posting there though!”
Go here:
link to facebook.com
Thanks guys, lots of useful stuff there.
Ah now, you see, that’s kind of proved that such jumps in likes can actually be legit as that silenced page has jumped from 22 likes to 34 in the 30 mins since you posted the link a more than 50% increase!
36
Apparently, you can boost your likes on FB if you pay for it. Someone was telling me recently but not too sure how to go about it. Although you can pay by Paypal.
Sorry late to this, re Facebook likes
A very dedicated poster on YTAIS (Craig Dalzell) has been tracking the daily likes of all 4 main pages for a few months now and shares a weekly update on the page, he would I’m sure confirm that the large spikes in BT numbers coincides with sponsored/paid links.
Here are last weeks updates:
weekly likes: link to fbcdn-sphotos-c-a.akamaihd.net
Daily likes: link to fbcdn-sphotos-e-a.akamaihd.net
Always enjoy seeing the steady ascent of the blue line weekly
Oh and Silenced by Better Together now at 349, no paid sponsorship…just alot of disgruntled folks sharing with each other.