At politicalowl, Richard Farmer quotes Gary Morgan on the Newspoll released today, which asked questions about the Prime Minister’s handling of asylum seekers but which also included questions about voting intention. But the results of those questions were not printed by its owners, News Limited:
The evidence was clear. Yet the publication of News Ltd’s poll (Newspoll) in the first place had already had a major impact. The evidence showing the ‘error’ of Newspoll was literally ignored by media discussion (e.g. the Insiders on ABC TV and the impact of the ‘rogue’ poll was allowed to run unabated).
Pollsters and those who publish the polls have a responsibility to report the facts and the truth.
Newspoll should have conducted another poll as soon as possible when they saw the dramatic change in their results — and if different, released the data to correct the misconceptions caused by their ‘rogue’ poll.
It is extremely worrying that today’s Newspoll on “boat people” clearly did include questions on ‘Political support,’ but the results from the ‘Political support’ question were not published.
A statistical analysis of the data reported on Australians’ attitudes to “boat people” issues — specifically the breakdown by ‘Political support’ — suggests the ALP vote in that poll was very strong. The percentage supporting each political party clearly should have been released.
Polls and their publishers should not seek to set the agenda by selectively releasing polling data.
Polls and their publishers are powerful but with that power comes responsibility.
Predictably tonight, the ABC and SBS news, and the 7 30 report ran with lines shaped by The Australian’s coverage of Newspoll, with no or just a bare mention of the Nielsen results. It may be that the voting intention results will be released tomorrow, but the delayed release and the lack of context to the results on questions about asylum seekers presents a picture which is deliberately distorted, stoking the claims about “crisis” and inflaming the issue further.
This really is getting to be a complete disgrace.
Update: William Bowe on Chris Mitchell’s explanation:
Queried by Andrew Crook of Crikey, The Australian’s editor Chris Mitchell explained that “even Crikey” should be able to understand that a non-fortnightly set of voting intention figures would cause a disturbance in the force. Mitchell further invoked a Beatles-and-the-Stones style arrangement between Newspoll and Nielsen in which they have agreed not to step on each other’s releases. Yet just one month ago, on the same day that Nielsen produced its regular monthly poll, The Australian published a “special Newspoll survey” on the Liberal leadership in between its regular fortnightly polls, and was not in the least bit shy about informing us that the sample produced the same 58-42 split in favour of Labor as recorded the previous week. In fairness, it should be noted that Crikey “understands that on Sunday morning, Newspoll chief Martin O’Shannessy contacted his Nielsen counterpart John Stirton and agreed not to release the two-party preferred vote to The Australian”.
From Morgan above:
“A statistical analysis of the data reported on Australians’ attitudes to “boat people” issues — specifically the breakdown by ‘Political support’ — suggests the ALP vote in that poll was very strong.”
Which, I presume, refers to this file:
http://media.theaustralian.com.au/multimedia/2009/11/newspoll.jpg
Which contains information such as [in table form]:
6% of the 1203 people polled answered “Very good” to the first question.
Of the unknown number/% of ALP supporters 9% of them answered “very good”.
And this format is repeated several times for 2 of the questions.
Is there any mathematical trick to discovering what the unknown number/% of ALP voters is [and ditto for COAL voters] from that sort of information?
Is that what Morgan is saying, that someone has done a ’statistical analysis’ and it suggests that ALP vote was ‘very strong”?
Does LP know anyone capable of emulating that trick and telling us the numbers?
So we can be let into Ltd Newspoll’s little secret?
For mine “fredex” at Possumland just about sums it up…
“I have an opinion on coffee, I like it, and tea, I don’t like it, whereas my wife is exactly the opposite. Strangely enough in neither case does our opinion on tea or coffee effect which political party we support. And it seems this is much the same effect that the boat people media-hyped issue has on the Australian public.”
You’re over-reacting Mark. The Newspoll was an aberration, the possibility of which had been acknowledged by Newspoll and a host of other commentators – including your good self – since day one.
Morgan is Newspoll’s commercial rival. Of course Gary Morgan – who has never produced a rogue poll in his life – is going to talk up Newspoll’s inadequacy. The next Newspoll is due out in one week’s time. I suspect that balance will be restored and normal transmission resumed, at that point.
You are NOT overreacting Mark, Geoff Honnor is totally wrong. Richard Farmer and about 2000 commenters on Pollbludger are correct – Newspoll is a perfidious disgrace, Disgraceful also was Barry Cassidy, the host of Insiders in the episode last Sunday morning. Cassidy, and the producers Kellie Mayo and Matt Brown should have been called into Mark Scott’s office and asked to explain themselves for promoting what is now a hoax.
Admittedly, the media loves what seems like a red-hot yarn: another Tampa. Except that it is all bullshit.
So why get hot under the collar if it is just a commercial rivalry? Because, Geoff Honnor, this is not a battle for shelf space between Marmite and Vegemite but at the heart of what people are thinking about the asylum seekers and consequently a political threat to the government. If one wanted to change government policy, one effective way would be to let the ferret run with a “rogue” poll. I don’t want to seem overly paranoid, and as Ambigulous sensibly pointed out in another thread, this could just be sloppy methodology.
Yet, for mine, the sloppy methodology defence is hard to credit. Newspoll are very experienced pollsters who have been doing this thing for many years, while the error seems ludicrously and amateurishly out of the ballpark.
The issue here is that the poll’s “findings” actually have the effect of altering public opinion. This is sinister and anti-democratic. People are intrinsically herd animals and tend to go with what is put about as the “majority” opinion. Very few want to be on the outer. There is even an Australian term for people with contrary views – ratbag. But it is evident from history that manipulation of public opinion by the media has changed public opinion to be in line with what the nexus of powers that be and powerful media wanted. History is full of examples, from the Britain of 1914 and Germany under Goebbels to steady chipping away at the public opinion regarding anthropogenic global warming – the conversation with the pollsters on last night’s 4 Corners was chilling and thought provoking.
We should all be alert for strokes like this. The price of freedom is eternal vigilance.
I think what Mark is getting at Geoff is that a new issue has arrived on the doorstep.
Namely the credibility and integrity, or otherwise, of Ltd Newspoll and those who feed off it. Endlessly.
I would be surprised however if the media pundits recognized it as such.
I was up early this am to work in the garden outside [by 8am it was 35 degrees here, climate change ya know] when I went outside ABC National was banging on about boat people, again, and when I came back in a hour or so later, they were still banging on about boat people.
Meanwhile in the real world other issues await.
Hannah’s Dad @ 1:
Can’t do it. We have a single equation with two unknowns:
9 x L + Z x (100 – L) = 6 x 100
where L is % ALP voters
and Z is % of non ALP voters who responded ‘very good’
Ta Jenny, your knowledge vastly exceeds mine.
So I wonder what Morgan was referring to in this:”A statistical analysis of the data reported on Australians’ attitudes to “boat people” issues — specifically the breakdown by ‘Political support’ — suggests the ALP vote in that poll was very strong.”
Oh and ABC National are still banging on about you know what via their news bulletin, quoting some bloke named Joe [not the smart fella here known as joe2].
Newspoll’s best interests would be served by publishing the new voting intention figures, given that the other pollsters are all in agreement. Every day that the new figures are held back is another day that Newspoll’s industry reputation is damaged, and perceptions of political bias at Newspoll is reinforced.
There would be a few disgruntled employees at Newspoll right now.
So, is Neswpoll the local version of Zogby, or Rasmussen?
If it’s the former, then we won’t have to worry about them much longer, as they’ll just fade away.
It’s got to be the latter, I suppose.
Why are people surprised that an organisation linked to Murdoch’s Australian turn out to be a bunch of shameless liars. Murdoch pulled crap like this in 1975 and he’s doing it again. I wonder what Rudd said no to him about?
To call it ‘Limited News’, as many do, is looking more and more justified, especially in light of this dodgy-looking episode. It seems to be not a case of publishing “all the news that’s fit to print”. Rather, it’s Limited to the News that helps their running dogs the Liberals to damage the Labor government.
And it’s not just them. Much of the mainstream media, including their (Liberals’) ABC, are working in chorus, doing the Liberals’ job for them, keeping the dogwhistling going with more than a month’s breathless beating up of this asylum seeker (non)crisis.
Well, I’m over it! The sooner this mob put up their paywall the sooner I can vote with my wallet not to support them in their partisan agenda.
Sir Henry, I entirely agree with your comment. These ratbags need to be exposed for what they are. I challenge and shame my friends and acquaintances for ‘collaborating’ with News Corp. Most respond by saying there is nothing else available and it is convenient. To which I respond “Thats what they say about fast food and look what it has done to our health”. In my opinion a bit of home cooking with selected and genuine ingredients is more fun and wholesome in the long run.
My malicious side would answer Paul Burns@10 that he could consider the tie up between Murdoch, Foxtel, and the possibility of Telstra having to sell off it’s half share in pay-tv if Telstra refuses to split it’s business as Conroy has threatened. However I have my doubts that the Newspoll organisation rigged the poll in question. What is not in doubt is the dishonesty involved in not publishing at this time the full figures which they apparently have. Notwithstanding this I’m still unclear why Martin O’Shannassy predicted on the day the rogue poll was published that it represented a trend. Surely one can only judge a trend after more than one poll event?
More News than Poll, huh? Hey, its your credibility funeral, News.
“We understand Newspoll, cos we PWN it!”
Trackback:
http://www.smh.com.au/news/opinion/polls-apart-at-the-sausage-sizzle/2007/07/13/1183833770404.html?page=3
Interesting Lefty E. Sad to say there’s no one left with the passion, courage and wit of Alan Ramsey.
I agree with Casingbrooke @4.
This rogue poll is an attempt to change public opinion and government policy.
Remember Rupert’s proud boast that Australians have elected the Prime Minister that Rupert wants [every election since 1960].
Geoff see the Spiral of Silence for more of an idea of what Sir Henry is talking about. It’s a big deal, this is corrupt – either morally or systemically, possibly both.
Apart from commenting on threads like this or on newspaper sites and the ABC about this disgravceful injustice what else can one do. (I’ve already stopped buying Murdoch papers.)
Rx @ 11, I recall back in the day that one of the underground papers (may have been Rolling Stone, but I’m willing to be corrected) had, as its masthead, a conscious parody of one of the major respectable newspapers who were busy lying about the Great Military Adventure in South East Asia, “All the news that fits.”
Sorta fits Limited News, doesn’t it?
Hannah’s Dad at 1: I think Labor’s share of the vote can be worked out from the table “Best to handle asylum-seeker issues” which gives Labor and Coalition percentages for five categories.
These categories are: Labor 21%, Liberal-National Party 22%, Someone Else 10%, Neither 13% and Uncommitted 34%.
Going across the column, we find 36% of Labor and 4% of L-NP make up the 21% for Labor. This works out at 18.9% for Labor and 2.1% for L-NP of the total 21%
For the L-NP, the 8% for Labor and 54% for the L-NP represents 2.83% for Labor and 19.16% for the L-NP of the total 22%.
For Someone Else, the 11% for Labor and the 4% for the L-NP works out at 7.33% for Labor and 2.67% for the L-NP of the total 10%.
For None, the 15% for Labor and the 9% for the L-NP represents 7.8% for Labor and 5.2% for the L-NP of the total 13%.
For Uncommitted, the 30% for Labor and 29% for the L-NP represents 17.3% for Labor and 16.7% for the L-NP for the total 34%.
Adding up the Labor total gives 54.16% and adding up the L-NP total gives 45. 83%.
The way the percentages are worked out is that in the first column 36 is 90% of 40 which corresponds to 18.9% of 21%. In the second column 8% of 62 is 12.9% of 62, which corresponds to 12.83% of 22 and so on.
Related to this “OMG-teh-Government-is-DOOMED-coz-of-asylum-seekers” meme that is being pushed at present, I was amused to hear this story over the airwaves this morning.
Obviously, because a 19yo student was allowed to speak her mind at a community event, TEH GUBERMENTS DAYS ARE NUMBERED!!!!
Or something. o.O
John L
Thanks for that.
I tried a similar sort of approach last night when I couldn’t sleep cos of the heat here.
I then realized there was a flaw or two in my thinking [which I have since forgotten, maths is not my strong point] and I also had no way of accounting for the Greens and “other’ categories,the ‘refused to answer” sector, if they exist as separate entities. We are not told how those 2 groups responded.
So I dunno.
I decided to ask an expert and Jenny above reckons it can’t be done.
Yet Morgan strongly hints something can be worked out.
Strange isn’t it?
“Geoff see the Spiral of Silence for more of an idea of what Sir Henry is talking about.”
Noted patrickg. Thanks. I’m still inclined to apply Occam’s razor rather than a conspiracy theory. Let’s see what happens next Tuesday, I reckon the ’spiral of silence’ will be last weeks news shortly thereafter.
I agree-ish with you Geoff – in that Newspoll, Nielsen and Morgan have, I believe, an understanding not to cannibalise each others’ results by publishing at the same time.
However, Newspoll would no doubt have been cognizant of the fracas, just as News Ltd were, and cognizant as to why News Ltd would be reluctant to get any info on 2PP. Really, it either behooves Newspoll to publish the information, or not to solicit it in the first place. To ask one question, whilst allowing your publisher to dodge the answers to the question that everyone actually wants to know is pretty frigging remiss, in my opinion. Also, you can be damned sure that if those results were negative for Rudd, the OZ would be hiring sky-writers.
I think Rudd might have to worry seriously ewhen the polls say Labor is not doing enough to combat global warming. Disgraceful though his behaviour has been over Tamil refugees both with the Oceanic Viking and Alex’s boat, I don’t really think Newspoll is really going to be able to get people into enough of a lather about it that they’ll start voting Liberal. Especially after last night’s 4 Corners. If ever there was a programme that would scare the horses, that was it.
hiring skywriters?
Why? Is nobody reading their newspapers ?
“Also, you can be damned sure that if those results were negative for Rudd, the OZ would be hiring sky-writers.”
No need when you have Their ABC on-side.
Ambigulous wins the meat tray.
Hannah’s Dad at 23: While they do not give figures for Greens or Independents,
the main possibilities are that they have excluded these figures and are showing only the percentages of the reduced figures (there were 1203 surveyed), or they have used all the figures and made some adjustment to the percentages of the ALP and l-NP based on past figures.
However, whichever way it is, this method uses their claimed percentages to work out a total percentage for Labor and the L-NP. A result of Labor 54.16 and LNP 45.83 in a poll on an issue that should not favour Labor is extraordinary.
I think the method I used (or something similar) is how Morgan worked out that the ALP percentage was high.
I used a couple of other approaches and had similar results.
“No need when you have Their ABC on-side.”
i don’t know, adrian. I reckon their ABC did a fine shop of shifting the debate for the government last night. Haven’t heard much about Newspoll or asylum seekers anywhere much today except here. Nick Minchin’s suicide bombing of the Coalition has had saturation coverage. The “spiral of silence” must be on the blink.
Here’s a simple question that I dare Newspoll to canvas:
Do you give a toss about Newspoll? Yes or No.
JohnL
I am half convinced by your numbers but lack the expertise to recognize what I am looking at, I have a nagging suspicion there is more to the problem than I can see.
Thanks for the responses.
Having no in-depth understanding of polling but a keen interest in politics I tend to rely on the Pollbludger and Possum before accepting any particular poll result. I would have assumed that journalists and those working in the MSM in general would have done something similar before launching into print or giving voice on air about a poll result so dramatically out of the ordinary. I have been astonished at the MSM’s complete ignorance or ignoring of such informed comment to say nothing of barely mentioning the published results of other pollsters like Nielsen and Morgan. And where has Anthony Green been since his comment on November 2nd which cast some doubt on that Newspoll and made passing reference to others which gainsaid it. And who is researching and writing for the Insiders, Lateline, 7.30 Report and Radio National News?
Surely it’s time for Rudd to act on the ABC as Howard did and to hell with being even handed. It’s seems we have a national broadcaster with a news and current affairs department which is either lazy and inefficient or biassed and subversive of our democracy.
Rudd by the way seems to have already got across the polls and informed comment on them, or more likely has his own internal polling to go by, judging by his consistent message on asylum seekers in general and refusal to budge on the Oceanic Viking issue. The MSM were delighted at his seeming initial panic and rush to the airwaves, but he hasn’t changed his message, just seems determined to get it out and to tough it out. I’m trusting his reputation as a consummate politician here and hoping his gut reaction to hold firm is the right one.
Asylum Seeker ? – the liberals didn’t have a clue when this situation first emerged – so the political commentators took over leading the argument and encouraging the opposition to engage – why – for the sake of a national debate presumably and knowing this was an issue that can divide the Australian community – good material for the media – I watched this debate emerge from day one and the media’s meddling in this has some democratic justification when you have such a intellectually weak opposition – however some commentators have crossed the line by inflaming the issue. Media should report on – not push for.
I wonder if Rudd is applying a little bit of the strategy that this blogger, at Anderson360, highlights about Obama:
http://ac360.blogs.cnn.com/2009/11/08/dear-president-obama-293-gone-fishing/#comments
As we can all recall from the endless Howard-staging-comeback propaganda at the last election, The Australian’s interpretation of Newspoll is usually highly suspect.
Didn’t we all love the creative ways The Australian could find the tiniest sliver of hope for the Tories in the bleakest of numbers? Meanwhile, any drop for Labor (usually within the margin of error) is a massive drop back to earth. On the slimmest of evidence The Oz has concluded that regional voters has abandoned Labor en masse – and it’s all due to the CPRS!! How do they know the CPRS is a burning issue out in the sticks? Nothing more than a hunch really – didn’t stop a week’s worth of commentary, though.
News Ltd has been desperate to find a “narrative” to use against Rudd – strippers, Brian Burke, air hosties, utes, debt-and-deficit, the horrible thought that there is a pink bat being wasted on a one-student school somewhere, refugees.
So far, touch wood, nothing seems to have stuck. It’s almost as if the electorate has gotten wise to all News Ltd’s bag of tricks.
…has? I meant have.
@25 – yep, I think the “competing pollsters in the field” explanation is a bit disingenuous, to put it charitably. You don’t have to have your conspiracy theory hat on to work out the big holes in the justification given.
You probably do have to have your conspiracy hat, Mark, to read “big holes in your justification” as some sinister plot.
Rupert published a graph poll favouring the ALP upside down, during an election campaign in SA back in the early 70s, from memory, so that it appeared to favour the Libs. So he is well capable of and willing to, tamper with polls. Rupert Murdoch and the truth are barely on nodding terms.
I noted the other day, Geoff, that Glenn Milne, with his usual lack of subtlety was banging on about how Newspoll – even if “wrong” was “real”. They’re their own best evidence as to the fact that they see their role as actively shaping not just reporting political events.
What one judges as a “sinister plot” is, of course, another question. I don’t doubt that the sample in the outlier was just out of whack, but how they respond to this, and how they work to impose one interpretation is pretty clear. It doesn’t require positing smoke filled rooms or whatever. It’s plain to see.
Update: William Bowe on Chris Mitchell’s explanation:
Does the ACCC know about this collusion?
Mitchell further invoked a Beatles-and-the-Stones style arrangement between Newspoll and Nielsen in which they have agreed not to step on each other’s releases
Does the ACCC or ACDC know about this collusion?
And nobody told us.
And nobody told us neever.
horror show there, right there on my TV …. it’s the 6.30 news
*looks pained*
*wanders off, dragging chains*
I’d come and eat your brains but from that comment @49, there wouldn’t be much there for me.
As discussed on Poll Bludger, could someone please explain why ABC news is still rabbiting on about the asylum seeker issue for all it’s worth, while apparently completely ignoring the Amanda Vantstone links to the Italian mafia story, that you would thinkany impartial news service would at least cover.
Because, adrian, as I understand it, the Mafia bloke got his visa because some people in the Mafia in Italy were not happy with him and would have done him some serious damage if he stayed in Italy? That info. is by the way, courtesy of our ABC2 breakfast program yesterday or the day brefore, I think.
OTOH, I too am very very annoyed about the ABC pushing the asylum seeker thing ad nauseam, but if Ministers are going to keep on giving interviews on the topic what is the ABc to do?
This is how Aunty buried the new Newspoll on it’s site this morning that proved conclusively the last one was crap.(It’s down the end of a totally unrelated report).
Note how they are still rabbiting on about asylum seekers even when there is no evidence it hurt Labor at all or they might have been party to spruiking a dodgy poll, last time, with that as its cause.
http://www.abc.net.au/news/stories/2009/11/17/2744532.htm
ABC coverage of the new poll was indeed very muted. When is K.rudd going to put a dose of salts through the place. After all, Ratty had no compunction stacking the board. Time for Goody-Two Shoes to get his act together.
)
Oh, and if the current Newspoll doesn’t prove the last one was a Goebbels-like lie, nothing does. (Have I broken Godwin’s Law?
Smashed it I think. The Newspoll beatup wasn’t in the Goebbels big lie class, though it might have lived in the next suburb.
It was more a combination of sloppy technique and overexcited journos playing into a well-established narrative: Aussies hate boat people. And let’s face it –look what happened when the indigenes let in those boat people in 1788.
FB @ 56,
Ah. 1788. The indigenous population on the shores of Botany Bay and the cliffs of Port Jackson shook their spears and shouted “Worra, woora, woora”, which politely translated by Captain John Hunter, was supposed to mean “Go away!” Personally, I reckon it meant “Fuck off!”
Funny how this current crop of boat people are behaving exactly like the first lot and ignoring the people who tell them they’re not welcome.
“…..and ignoring the people who tell them they’re not welcome.”
If I were a cartoonist I would have Turnbull, Sharman Stone and Barnaby in the front row, decorated in leaves with blackface, waving spears at the approaching asylum seekers.
At the back encouraging them, from a safe place, would have to be Dolly and Johny in an appropriate loincloth and stockings for the former, of course.