-
Categories
-
Pages
Politics, elections and piffle plinking
-
-
Politics, Media, Business ... No Britney
Read the Stories you won't get anywhere else.
-
-
-
NSW Polling
-
QLD Polling
-
VIC Polling
-
SA Polling
-
WA Polling
-
Pollbludger -
Antony Green
-
Larvatus Prodeo
-
The Tally Room
-
Meganomics
-
Club Troppo
-
Core Economics
-
Jack the Insider
-
Peter Martin
-
Pure Poison
-
Andrew Bartlett
-
Woolly Days
-
Political Sword
-
Public Opinion
-
Tasmanian Politics
-
First Blog on the Moon
-
OzPolitics.info
-
Copyright © 2008 Private Media Pty Ltd, Publishers of Crikey.com.au. All Rights Reserved.
Independent news, blogs and commentary on politics, media and business.
Editor: Jonathan Green.
Publishers: Eric Beecher, Diana Gribble.
Level 7, 22 William St, Melbourne, 3000. Ph: 1800 985 502 Fax: (03) 8623 9975
No Comments
Pages: « 1 2 [3] Show All
Just on the “how did you vote at the last election” question, my recollection was that the longer you leave it, the more people misremember their vote as one for the winning party, if they weren’t highly committed to the choice at the time. The classic instance of this is after Kennedy’s assassination, when I think something like 65 or 70% of a national sample claimed to have voted for him, when, as everyone knows, there was only a cigarette paper’s worth (or a few stuffed ballot boxes in Chicago and Texas) between Nixon and Kennedy. That’s an extreme case, but I’m pretty sure that I remember reading that it’s a phenomenon of usual occurrence.
Just on the “how did you vote at the last election” question, my recollection was that the longer you leave it, the more people misremember their vote as one for the winning party, if they weren’t highly committed to the choice at the time. The classic instance of this is after Kennedy’s assassination, when I think something like 65 or 70% of a national sample claimed to have voted for him, when, as everyone knows, there was only a cigarette paper’s worth (or a few stuffed ballot boxes in Chicago and Texas) between Nixon and Kennedy. That’s an extreme case, but I’m pretty sure that I remember reading that it’s a phenomenon of usual occurrence.
It poses an interesting question Mark. On the one hand I’m not a very big believer in after the fact polling on events, which is why I rarely touch the Australian Election Survey results, but on the other hand, such a question would be really helpful for political analysis even if it was a bit biased toward the incumbent.
I’ve often wondered how much of that post-event bias is a result of the faulty memory of survey respondents and how much is as a result of respondents agreeing with the majority after the fact. It would be interesting to test two different methodologies on this question with two separate groups - where the first group was simply asked who they voted for at the last election, while the second group would be asked to answer the same question but instead of doing it verbally, were asked to enter their answer via the keypad on the phone (1 for ALP, 2 for Coalition, 3 for Others) and the person asking the questions let the respondent know that they couldn’t see their answer.
I bet London to a brick on that the answers of the second group would be closer to the actual election results than the first.
It poses an interesting question Mark. On the one hand I’m not a very big believer in after the fact polling on events, which is why I rarely touch the Australian Election Survey results, but on the other hand, such a question would be really helpful for political analysis even if it was a bit biased toward the incumbent.
I’ve often wondered how much of that post-event bias is a result of the faulty memory of survey respondents and how much is as a result of respondents agreeing with the majority after the fact. It would be interesting to test two different methodologies on this question with two separate groups - where the first group was simply asked who they voted for at the last election, while the second group would be asked to answer the same question but instead of doing it verbally, were asked to enter their answer via the keypad on the phone (1 for ALP, 2 for Coalition, 3 for Others) and the person asking the questions let the respondent know that they couldn’t see their answer.
I bet London to a brick on that the answers of the second group would be closer to the actual election results than the first.
Yes, Possum, I agree it would be very helpful. The meaning of “political affiliation” in this context isn’t particularly clear either though - though presumably it’s a crosstab with the voting intention question, you could get something more meaningful if it was tabulated with the intensity of the voting choice. Does Newspoll push people to make a choice? We used to in some ALP polling we did yonks ago - in the sense of asking people who they were leaning to, if they would change easily, etc - for tracking polls.
I’ve got a feeling actually that my recollection of this problem actually comes from some methodological discussion from an old AES.
Yes, Possum, I agree it would be very helpful. The meaning of “political affiliation” in this context isn’t particularly clear either though - though presumably it’s a crosstab with the voting intention question, you could get something more meaningful if it was tabulated with the intensity of the voting choice. Does Newspoll push people to make a choice? We used to in some ALP polling we did yonks ago - in the sense of asking people who they were leaning to, if they would change easily, etc - for tracking polls.
I’ve got a feeling actually that my recollection of this problem actually comes from some methodological discussion from an old AES.
Mark,
Newspoll does the gentle push on voting intention, asks people who they would vote for and if uncommitted, asks them which party they are leaning towards. Usually between 4% and 6% of the sample are removed from the voting intention counts because they remain uncommitted or simply refuse to answer.
I have sympathy for party pollsters when trying to get a handle on partisan intensity or vote certainty, as an awful lot of people see themselves as swinging voters but in reality just aren’t. In extraordinary circumstances these folks would change their vote - but circumstances are rarely extraordinary and trying to weed those out from the real soft and swinging voters is quite an art.
And as you know, it’s expensive to do this stuff properly. Just after the election I was approached by a group looking to start up the perfect political poll - a carefully selected sample of 1000 people used for a longitudinal political and public affairs poll undertaken once a month. To nail it down properly in a way that would provide high levels of certainty would have cost about $1.5 million a year to run. About 3 times more than they were willing to pump into it.
It’s a pity - she would have been formidable and have changed the way political analysis is done in Australia.
We’ll probably have to wait to Yougov moves into Australia to get anything approaching it - but even then, without a major media supporter to publish and spruik the results it becomes commercially questionable from the start.
Mark,
Newspoll does the gentle push on voting intention, asks people who they would vote for and if uncommitted, asks them which party they are leaning towards. Usually between 4% and 6% of the sample are removed from the voting intention counts because they remain uncommitted or simply refuse to answer.
I have sympathy for party pollsters when trying to get a handle on partisan intensity or vote certainty, as an awful lot of people see themselves as swinging voters but in reality just aren’t. In extraordinary circumstances these folks would change their vote - but circumstances are rarely extraordinary and trying to weed those out from the real soft and swinging voters is quite an art.
And as you know, it’s expensive to do this stuff properly. Just after the election I was approached by a group looking to start up the perfect political poll - a carefully selected sample of 1000 people used for a longitudinal political and public affairs poll undertaken once a month. To nail it down properly in a way that would provide high levels of certainty would have cost about $1.5 million a year to run. About 3 times more than they were willing to pump into it.
It’s a pity - she would have been formidable and have changed the way political analysis is done in Australia.
We’ll probably have to wait to Yougov moves into Australia to get anything approaching it - but even then, without a major media supporter to publish and spruik the results it becomes commercially questionable from the start.
Yep, it’s a pity!
Yep, it’s a pity!
Some of the stuff I deal with involves subjective retrospective recall in medical patients, often of events that occurred decades previously, and it is a well known nightmare trying to get reliable answers.
People are not being dishonest or evasive and are genuinely trying to remember accurately, but it is just the way human memory works, it is reconstructive and imbued with all sorts of post factum influences (emotional, intentional, interpretative, rationalisations, not wanting to go against the crowd, etc) and is most certainly is not a simple factual recall.
The problem is so great, that generally speaking, in medical science retrospective studies are not considered strong evidence of factual history, prospective studies are given far greater weight.
The study you suggest in #19 could be very interesting, Possum.
Some of the stuff I deal with involves subjective retrospective recall in medical patients, often of events that occurred decades previously, and it is a well known nightmare trying to get reliable answers.
People are not being dishonest or evasive and are genuinely trying to remember accurately, but it is just the way human memory works, it is reconstructive and imbued with all sorts of post factum influences (emotional, intentional, interpretative, rationalisations, not wanting to go against the crowd, etc) and is most certainly is not a simple factual recall.
The problem is so great, that generally speaking, in medical science retrospective studies are not considered strong evidence of factual history, prospective studies are given far greater weight.
The study you suggest in #19 could be very interesting, Possum.
I vaguely recall, but can find no trace using basic research tools, that something similar was attempted back in the early/mid 70’s, where respondents were asked to nominate which party they had voted for at the 1972 election. The question was randomly varied by asking if respondents had voted for McMahon or Whitlam. An improbable result ensued when most respondents declared they had not voted for the named leader, ie that few wanted to admit that they voted for leaders who had subsequently become damaged goods (or in McMahons case remained damaged goods).
I vaguely recall, but can find no trace using basic research tools, that something similar was attempted back in the early/mid 70’s, where respondents were asked to nominate which party they had voted for at the 1972 election. The question was randomly varied by asking if respondents had voted for McMahon or Whitlam. An improbable result ensued when most respondents declared they had not voted for the named leader, ie that few wanted to admit that they voted for leaders who had subsequently become damaged goods (or in McMahons case remained damaged goods).
Nightwatchman on 7:30 report - all at sea trying to justify the Lib policy… he was chewing the shit sanger vigorously… but it is the viewer who found it hard to swallow.
Nightwatchman on 7:30 report - all at sea trying to justify the Lib policy… he was chewing the shit sanger vigorously… but it is the viewer who found it hard to swallow.
Poss surely the 15% ALP less likely is irrelevent? I think your comment that he wont pick up votes is incorrect.
If 15% of ALP voters are less likely to vote LNP, that is irrelevant to the result, as nothing will change.
What you need to look at is the break down by party of who is going to change.
If switching to Costello results in 15% of ALP voters being likely to change to LNP, and 5% of LNP voters are likely to change to ALP, then switching to costello netts them a 10% gain.
The less likely number is irrelevant, because that only states that voters will stick with their current choice, thats not what we are interested in knowing.
I think you’ve concentrated on the wrong numbers.
Poss surely the 15% ALP less likely is irrelevent? I think your comment that he wont pick up votes is incorrect.
If 15% of ALP voters are less likely to vote LNP, that is irrelevant to the result, as nothing will change.
What you need to look at is the break down by party of who is going to change.
If switching to Costello results in 15% of ALP voters being likely to change to LNP, and 5% of LNP voters are likely to change to ALP, then switching to costello netts them a 10% gain.
The less likely number is irrelevant, because that only states that voters will stick with their current choice, thats not what we are interested in knowing.
I think you’ve concentrated on the wrong numbers.
Possum Comitatus Says:
July 29, 2008 at 12:04 pm
Alastair - how right you are!
Ideally the pollsters would add 1 single question to their polls: “Who did you vote for at the last election” and most of that could be answered.
<<
A most elegant solution: truly!
Let’s spam the idea and see if it takes off, shall we!?!
Possum Comitatus Says:
July 29, 2008 at 12:04 pm
Alastair - how right you are!
Ideally the pollsters would add 1 single question to their polls: “Who did you vote for at the last election” and most of that could be answered.
<<
A most elegant solution: truly!
Let’s spam the idea and see if it takes off, shall we!?!
Possum Comitatus Says:
July 29, 2008 at 8:47 pm
Spelling Nannas unite!
Ta folks.
Did anyone catch the Nelson media scrum late this afternoon?
Most of Nelsons silliness is self inflicted, but watching him try to make that shit sandwich sound as if it tasted like chicken was a real cringe moment. With the Libs inflicting so many of these problems on themselves, surely it must bring into question the political nous of the people surrounding Nelson, like Hendy.
And what I dont get is just who was rolled at this shadow cabinet meeting since they ended up back at the same position they had early last week?
<<
Jack the Insider said it was Minchin who got rolled and, apparently, is losing influence : but not having followed the story(I usually spend my time fighting in the comments section) I’m not quite sure if that adds up at all.
Possibly, tho, given Jack seems to give a bit to both sides!
Possum Comitatus Says:
July 29, 2008 at 8:47 pm
Spelling Nannas unite!
Ta folks.
Did anyone catch the Nelson media scrum late this afternoon?
Most of Nelsons silliness is self inflicted, but watching him try to make that shit sandwich sound as if it tasted like chicken was a real cringe moment. With the Libs inflicting so many of these problems on themselves, surely it must bring into question the political nous of the people surrounding Nelson, like Hendy.
And what I dont get is just who was rolled at this shadow cabinet meeting since they ended up back at the same position they had early last week?
<<
Jack the Insider said it was Minchin who got rolled and, apparently, is losing influence : but not having followed the story(I usually spend my time fighting in the comments section) I’m not quite sure if that adds up at all.
Possibly, tho, given Jack seems to give a bit to both sides!
It is pretty clear who got rolled - it was Nelson (and the people he is taking advice from) - he wanted to alter the policy and realised he didn’t have the numbers to do it….The press conference was akin to watching a death roll….
It is pretty clear who got rolled - it was Nelson (and the people he is taking advice from) - he wanted to alter the policy and realised he didn’t have the numbers to do it….The press conference was akin to watching a death roll….
Deathridesahorse, I didn’t catch the media scrum after yesterday’s LNP Party meeting, but I did see Brendan Nelson’s gibberish on the 7.30 Report. It was a wonder to behold. Kerry O’Brien became almost speechless with exasperation trying to pin him down. Then Greg Hunt, who always looks like a boy on a man’s errand, appeared on Lateline. Looking like an eager-to-please schoolboy, he had his new lines well-rehearsed and was determined to get them out. All that did was to exasperate Leigh Sales who felt the need to preface the repeat of an unanswered question with “Greg Hunt, Greg Hunt, Greg Hunt…”, Undeterred, he pressed on with that irrepressible smile he hopes will conceal his muddled presentation.
For sheer entertainment, such performances are almost as good as Clarke and Dawe; for information transfer to the general public, they’re inept and counterproductive. If the Coalition had deliberately set out to create confusion about its climate change policy it could not have done a more proficient job. The further it goes the worse it gets. By now I expect it’s only the political tragics who are paying attention. The main players would be wise to give it a break for a week or two, let the confusion subside and start again, preferably with just one unchanging message from all who speak on the subject. Perhaps they should take a leaf out of Malcolm Turnbull’s book; because he’s consistently uncomfortable talking on climate change, he avoids the debate whenever he can and thereby minimizes the opportunity to put his foot in it.
Deathridesahorse, I didn’t catch the media scrum after yesterday’s LNP Party meeting, but I did see Brendan Nelson’s gibberish on the 7.30 Report. It was a wonder to behold. Kerry O’Brien became almost speechless with exasperation trying to pin him down. Then Greg Hunt, who always looks like a boy on a man’s errand, appeared on Lateline. Looking like an eager-to-please schoolboy, he had his new lines well-rehearsed and was determined to get them out. All that did was to exasperate Leigh Sales who felt the need to preface the repeat of an unanswered question with “Greg Hunt, Greg Hunt, Greg Hunt…”, Undeterred, he pressed on with that irrepressible smile he hopes will conceal his muddled presentation.
For sheer entertainment, such performances are almost as good as Clarke and Dawe; for information transfer to the general public, they’re inept and counterproductive. If the Coalition had deliberately set out to create confusion about its climate change policy it could not have done a more proficient job. The further it goes the worse it gets. By now I expect it’s only the political tragics who are paying attention. The main players would be wise to give it a break for a week or two, let the confusion subside and start again, preferably with just one unchanging message from all who speak on the subject. Perhaps they should take a leaf out of Malcolm Turnbull’s book; because he’s consistently uncomfortable talking on climate change, he avoids the debate whenever he can and thereby minimizes the opportunity to put his foot in it.
Could that possibly be it?
Could that possibly be it?
Come on guys, Labor are done for…. according to the Oz Nelson is on the rise:
“Mr Rudd’s perception as being in touch with voters took a heavy hit, falling 13 points to 73per cent, while Dr Nelson’s rose five points to 44 per cent.”
Reality anyone?
Come on guys, Labor are done for…. according to the Oz Nelson is on the rise:
“Mr Rudd’s perception as being in touch with voters took a heavy hit, falling 13 points to 73per cent, while Dr Nelson’s rose five points to 44 per cent.”
Reality anyone?
Given public opinion you would think the LNP would have enough sense to adopt an acceptable or even more innovative position, knowing that if they ever won an election they could then manipulate it to suit their scepticism and hide the fact with obtuse words (Im sure the press would help them). Howard was good at saying he was doing some such think when what he was doing had the opposite effect.
Their problem is they hate to see Rudd receive kudos for having a policy, the alternative for them is to be on the wrong side of public opinion.
Given public opinion you would think the LNP would have enough sense to adopt an acceptable or even more innovative position, knowing that if they ever won an election they could then manipulate it to suit their scepticism and hide the fact with obtuse words (Im sure the press would help them). Howard was good at saying he was doing some such think when what he was doing had the opposite effect.
Their problem is they hate to see Rudd receive kudos for having a policy, the alternative for them is to be on the wrong side of public opinion.
Ad AStra:-
>Then Greg Hunt, who always looks like a boy on a man’s errand, appeared on Lateline. Looking like an eager-to-please schoolboy, he had his new lines well-rehearsed and was determined to get them out. All that did was to exasperate Leigh Sales who felt the need to preface the repeat of an unanswered question with “Greg Hunt, Greg Hunt, Greg Hunt…”, Undeterred, he pressed on with that irrepressible smile he hopes will conceal his muddled presentation. >
I’m amazed at the way in which certain media commentators continually talk up Hunt as being one of the Coalition’s heavy-hitters. When he was named as Environment spokesman, said commentators brayed that he would wipe the floor with Peter Garrett, who (purely on the basis of a couple of well-publicised election campaign gaffs) they perceived as being a poor Parliamentary performer. Since then, Hunt seems to have barely laid a glove on Garrett in the House, and his public performances appear equally weak. Granted, he has to sell a total dog’s-breakfast of policy, and he does appear a nice enough bloke - but is this really one of the Liberals’ shining new lights?
Perhaps he could form half of a leadership Dream Team with Cossie? : )
Ad AStra:-
>Then Greg Hunt, who always looks like a boy on a man’s errand, appeared on Lateline. Looking like an eager-to-please schoolboy, he had his new lines well-rehearsed and was determined to get them out. All that did was to exasperate Leigh Sales who felt the need to preface the repeat of an unanswered question with “Greg Hunt, Greg Hunt, Greg Hunt…”, Undeterred, he pressed on with that irrepressible smile he hopes will conceal his muddled presentation. >
I’m amazed at the way in which certain media commentators continually talk up Hunt as being one of the Coalition’s heavy-hitters. When he was named as Environment spokesman, said commentators brayed that he would wipe the floor with Peter Garrett, who (purely on the basis of a couple of well-publicised election campaign gaffs) they perceived as being a poor Parliamentary performer. Since then, Hunt seems to have barely laid a glove on Garrett in the House, and his public performances appear equally weak. Granted, he has to sell a total dog’s-breakfast of policy, and he does appear a nice enough bloke - but is this really one of the Liberals’ shining new lights?
Perhaps he could form half of a leadership Dream Team with Cossie? : )
Which reminds me to ask many months later, did we ever come to a conclusion as to why the last Nielson before the election was so far off?
Umm. that’s looking suspiciously like a (shudder) Widening. I do love the preoccupation with changing Leaders of the Opposition, as if that’s going to make any difference, in the MSM and the polling. Marsupial fluff.
Snapper - Would anyone WANT to be leader of such a motley crew of brigands?
Robert - Nielsen probably had some systemic bias in their demographic weighting last year which drove them to be a few points out. It probably wasnt a rogue poll as their final phone poll matched perfectly with their final online poll and both polls had over 1400 respondents each (from memory), which suggests something more than random chance produced the problem.
All pollsters regularly update their methodology and weighting, especially after elections. ACN is an extremely professional outfit, they would have done that very thing.
Pages: « 1 2 [3] Show All