Possums Pollytics

Politics, elections and piffle plinking

Posts Tagged ‘newspoll’

The Coalitions’ Demographic Train Wreck

Posted by Possum Comitatus on July 6, 2008

Back in March over at Australian Policy Online, Ian Watson published a really interesting paper titled “Is demography moving against the Coalition? ”, which was an update of a larger, earlier paper that added new results for the 2007 election period. What Ian Watson did was use Newspoll figures to look at the way different age groups have been changing their voting intention patterns over the period of 1987 through to the present.

What is really interesting about this paper is the dataset it contains at the end – age profile breakdowns on primary voting intention going all the way back to 1987 when Newspoll first started.

It doesn’t take long playing around with the data to realise that, putting it bluntly, the Coalition is facing a demographic train wreck of catastrophic proportions. It isn’t some short term problem that just appeared at the last election and which could easily be dealt with by a bit of vote targeting. Far from it, the impending train wreck is the result of a long slow demographic assault on the Coalitions’ primary vote that has been happening for at least 21 years.

They are losing ground among all age groups under 60, their only strong voting age demographic, the pre-Boomer over 60’s, are declining in number through attrition and will start being replaced by more Labor oriented Boomers over the next decade. As we will see, this pre-boomer demographic is carrying a large weight of the Coalition’s voter support. When that vote becomes neutralised by boomers moving into the 60+ age group, which is expected to occur sometime around 2018-2020, the Coalition primary vote will have lost around 4 to 5 points, perhaps a little more, should prevailing long term trends continue for the next 10-15 years in the same way they’ve played out for the last 21 years.

First up, we’ll just repeat what Ian Watson did and show how various age group voting intentions have been running for the Coalition at each election period since 1987 to give us a bit of a feel for the data. This data is Newspoll data, so the average Coalition primary vote result will be slightly different from the primary vote result that they achieved at each election, simply because of sampling error and late movement, but not by a great deal – it works out as an average of 1.47% mean absolute error.

As there are quite a few age cohorts here, we’ll split the demographics into two groups; the under 40’s and the over 40’s and we’ll also add the average Coalition primary result as well to show which groups are under and over that average.

Notice here that all groups under 40 have had a voting intention less than the average Coalition voting intention for every election since 1987.

Here it starts to get interesting. Up to and including the 1993 election, all groups over the age of 40 supported the Coalition to levels higher than the Coalition average support. But in 1996, the 40-44s were voting under the average, in 1998 the 40-49s were voting under the average, by 2001 it became the 40-54s all voting under the average and by 2007, the 55-59 group was voting just slightly above average but will probably vote below average next election and beyond.

If we play around with the data a bit and subtract the average Coalition primary vote estimate from each age cohorts’ support level for the Coalition, it shows this in starker terms. So, for instance, if the Coalition average was 40% and a particular demographic had only 36% support for the Coalition, they’d get a score of -4.

Again, we’ll do it for both the under 40’s and over 40’s.

From these two charts we can see that it is the over 60’s that are really carrying the weight of Coalition average support here. As the levels of support for the Coalition decline in younger cohorts, it drags down the average Coalition vote, leaving the Coalition more and more reliant on that over 60’s group to shore up their vote. But the problem here is that the over 60’s group is just about to be flooded with Baby Boomers, which will start reducing the Coalition dominance in their most important age cohort.

We can see how this might play out if we use the Newspoll data to track how people born between certain years have behaved over the last 21 years. The problem we have with the data here is that the age groups we’ll track don’t perfectly fit into the age classifications we have – but we can get pretty close on a number of elections. For instance, if we track the 25-29 age bracket from the 1987 election onwards, in the 1993 election that group would be 31-35, but we don’t have that as an available cohort. Yet we do have the 30-34 which is only out by 1 year. At the 1998 election those people were 36-40 years of age, and we can use the 35-39 age cohort for that and so on an so forth.

In the following chart, each age group is only out by a maximum of a year either side of their actual age, so it’s a fairly decent match to give us an idea of how people born in different years have voted over time.

There’s a couple of things to note here. Firstly, I stopped tracking groups when they got into the 60+ age group because it contains too many different ages all bunched together to be useful, so our last age cohort we can use effectively is the 55-59 age bracket. Secondly, you’ll notice a big drop in Coalition support in 1998 by those born between 1938 and 1942. Most of that lost Coalition vote went to One Nation. In 1987 that group voted 7.3% for minor parties and independents, in 1993 it was 6.5%, but in 1998 it was a big 17.3%. In 2001 that vote would have jumped back up to 50+% for the Coalition.

As you can see, those early Boomers born between 1948 and 1952 vote for the Coalition in substantially less numbers than do the older age groups – yet this group has just started to turn 60 this year. By the next election nearly all of that group will be over 60, by the election after that, the 1953-1957 boomers will be starting to turn 60. As time goes on, the Coalitions hold on that over 60 demographic gets further washed out, especially since many of the pre-boomer Coalition supporters in the 60+ group will be increasingly dying out.

If we take these same age groups and do what we did before and measure the difference between the average Coalition vote and the Coalition vote for each age group we get:

We would expect those born between 1938 and 1947 to have increased their “difference from the mean” over the last few elections if we could actually measure it with the Newspoll data, but unfortunately we can’t. However, since that group is reducing in number every year at a faster rate than their younger cohorts, it is slowly allowing the average Coalition vote to fall, and as a result reducing the “difference from the mean” for the younger cohorts.

Looking back over all of the charts, the voters the Coalition are losing aren’t being replaced by younger voters, to the point where it’s reducing the total Coalition primary vote. If the trends that have been happening for the last 21 years continue for the next decade, by 2018 thereabouts, the ALP will simply become unbeatable with TPP results coming in with a an expected demographic floor of around 55%.

(I thought I better bold that point)

So the Coalition has to start appealing to much younger demographics or they will likely find themselves in permanent opposition.

Something for them to keep in mind if they start trying to play political games with the emissions trading system and climate change – issues with large support in the younger demographics.

I’ll do some more with this data later, as well as start applying these results on a seat by seat basis to see which regions face the largest political changes, but I thought you folks might be interested to see how your sub-generation has been voting over the last 21 years or so. I’ll also take a closer look at those born after 1969 in another post.


I’ll stick this chart in here as well since it’s relevant. It’s a basic graphic of the results of the two Newspolls that dealt with climate change and the proposed ETS.

The demographics that the Coalition needs to attract are the ones that have the strongest views on climate change, willingness to pay for it and the benefit of an ETS.

It makes the politicking of the Opposition a hard slog for any long term partisan benefit.

AddThis Social Bookmark Button

add to kwoff


Posted in Election Forecasting, Political Risk, Polling, Voting behaviour | Tagged: , , | 25 Comments »

Putting the Newspoll in perspective and US election updates.

Posted by Possum Comitatus on June 17, 2008

Mr Mumbles again donned his secret squirrel cape and has the good acorns on todays Newspoll over here:

Primaries are running ALP 46/33 leading to a TPP of 59/41.

But to throw all these polls in perspective – let’s chart every poll taken so far in 2008 (by Morgan, Newspoll and ACN) and run a Loess regression through it as a line of best fit.

What is really noticeable here is that the drop in the ALP primary vote over time has been greater than the drop in their two party preferred – and larger than the rise for the Coalition primary vote that started around day 110.

The Greens and “others” vote have been the beneficiaries of this falling ALP primary, letting it flow back the ALP in preferences for TPP terms.  As we found last year, we often get a bit of noise in the minor party vote changing the TPP headline number by a few points here and there, but currently we are getting small movements in the minor party vote that is keeping the TPP numbers where they are and changing the underlying primary vote composition of the polls.

To see this, we only have to look at the ALP primary vote in the context of both the Coalition and Greens+Others vote. To do this, we’ll chart the ALP vote as inverted (meaning it decreases as you go up the vertical axis on the left) and chart the Greens+Others and the Coalition Primary normally on the right hand side axis . To see where the votes are shifting to, if the lines move together, then votes are shifting between ALP and the other party on the chart, if they move in opposite directions then that isn’t happening.

Up until the beginning of May (the 3rd Newspoll), voters were moving between the ALP and the Coalition as well as between the ALP and the Greens minor parties.

For the following two Newspolls votes were only moving between the ALP and the minors, and finally in the latest poll, votes moved from the Coalition to the minor parties.

We can also see this playing out in the satisfaction ratings.If we look at the satisfaction ratings of Rudd vs Nightwatchman  and how they’ve changed over time, Rudd at the moment has his second lowest satisfaction rating recorded this year by Newspoll at 59%, while the Nightwatchman has satisfaction and dissatisfaction ratings that haven’t moved a jot. As voter dissatisfaction has increased for Rudd, votes aren’t moving to the Libs as a result, they’re moving to the Greens and preferences are flowing back to the ALP two party preferred.

If we chart both the ALP TPP vote and the ALP primary vote against Rudds satisfaction level, we get:

We expect satisfaction ratings and vote levels to move together, but the satisfaction rating is having a much larger influence on the ALP primary than the TPP.

This is because most of the change in the ALP primary vote is moving to the minor parties,- these voters might not be impressed with what Rudd is doing, but they are refusing point black to support the Coalition.

On the Coalition side, they just recorded their lowest primary vote of the year at 33% driven by some movement from them to the minor parties – which is a bit unusual and probably a sampling artefact rather then any true indication of a large change in Coalition primary vote support. But regardless, I cant imagine we’ll be hearing any more Honeymoon is over stories this week, let alone Rudd in danger of being a one term wonder from the shallow end of the commentariat pool.

In other news, the most accurate aggregation of polling around, our Pollytrack series currently has the ALP leading 56.9% to 43.1% in TPP terms,  off the back of primaries running 45.1%  to 37.4% to the ALP – all with a margin of error or 1.56% and a sample of 3938

Over at the Pollytics US Election page, all the Intrade data has been updated and now includes large Monte Carlo simulations to get a better idea of the probability spread on the Electoral College Votes by State, as well as cumulative frequency charts of these simulations to show how the probability of the Intrade market has changed over the last month for every electoral college vote number –worth a squiz if your nerdy.

Posted in Polling, Voting behaviour | Tagged: | 7 Comments »

State of the Polls – Old School

Posted by Possum Comitatus on May 4, 2008

It’s been an awfully long time since we last had a good look at the broad state of the political polling like primary votes, satisfaction ratings and their trends rather than the simple headline TPP and Brendan Nelson’s limbo with preferred PM ratings.

So to start off, some music to listen to as we go through the poll roll – the Halo Friendlies doing “Sellout”… quite apt for the poor old Nightwatchman of late.

Looking first at the primary votes of the two majors, we’ll plot how Newspoll, Morgan Phone Poll, Morgan Face-to-Face and the Phone Poll Average have been going since the election. The time scale at the bottom of all these graphs is “Week in Term” meaning week one was the first week after the election, week 2 the 2nd week and so on and so forth. To help convert the weeks in term into dates, this little table might come in handy:

Week Ending Week into term
22 December 2007 4
29 December 2007 5
5 January 2008 6
12 January 2008 7
19 January 2008 8
26 January 2008 9
2 February 2008 10
9 February 2008 11
16 February 2008 12
23 February 2008 13
1 March 2008 14
8 March 2008 15
15 March 2008 16
22 March 2008 17
29 March 2008 18
5 April 2008 19
12 April 2008 20
19 April 2008 21
26 April 2008 22
3 May 2008 23
10 May 2008 24
17 May 2008 25
24 May 2008 26
31 May 2008 27
7 June 2008 28
14 June 2008 29
21 June 2008 30

For the ALP we get:

The Morgan face to face poll went wandering out to the fringes of plausibility here for a bit but has lately started coming back to earth – essentially repeating what it did last year. What’s worth noting here is that the two phone polls have pretty much been moving in sync, with the old pattern of Morgan usually being a few points higher for the ALP than Newspoll coming through just like it did last year. Since the end of February there appears to have been a slight growth in the primary vote for Labor which, as we’ll see a little later on, is consistent with the way the uncommitted voters are splitting over other metrics like satisfaction and preferred PM ratings.

Next up, the Coalition primary vote:

Again, the Morgan Face-to-Face poll is the odd one out, being a fair bit more volatile than the phone polls, but also lately showing the highest primary vote for the Coalition. The minor party vote in Morgan’s face to face seems to be a good chunk less than the phone polls are measuring – one would think it’s a bit of a methodological issue going on there.

Using the phone poll average as the comparison between the two parties we end up with:

We can see some slow, consistent movement to the ALP in primaries over time, but only partially at the expense of the Coalition vote – with the rest coming from a decreasing minors vote.

Moving on to the TPP vote estimates – we’ll again compare the pollsters for the two majors.

Apologies for that dogs breakfast – blame the pollsters!

Worth a giggle is the Morgan face to face poll showing an ALP TPP of 65%. Were an election held where that result came about, the ALP would have 139 of the 150 seats in Parliament :mrgreen:

The ALP TPP vote seems to have been slightly growing over the last few months yet without being able to say so with any level of statistical significance.

So now we’ve done the primaries and the TPPs we can have a squiz at the Votegap – which is the difference between the vote levels of each party.

This must be more than just a little bit disturbing for the Coalition. As time goes by, the difference between the vote levels of the ALP and the Coalition in both primary votes and TPP share is increasing. That suggests that ALP support is not only coming from minor parties, but directly from the Coalition as well. The ALP seems to be incrementally grinding away into the Coalition base vote – which is exactly what happened in QLD state politics. If a party starts losing its base they are in deep shit.

One blessing for the Coalition is that there aren’t yet enough observations in the data to be able to say this with any real level of statistical certainty – but if it keeps happening like it is at the moment, by the time we get enough observations it will probably be too late for the Opposition to be able to recover before the next election which will put them in a dire position for the election following that – especially in terms of resources.

We’ll keep an eye on the Votegap measures over the next year and see if that longer term trend to the ALP continues – however slowly. If it does, Australian federal politics will change fundamentally.

Moving on to the more qualitative metrics, we’ll start off having a squiz at Nelsons limbo dance with the preferred PM ratings:

This has just about been done to death in the MSM so we’ll leave it pretty much alone except to say that as the uncommitteds have started to crystalise out and get an opinion, they’ve clearly decided that they don’t much think Nelson is up to the job.

That uncommitted pattern is also something worth looking at in the satisfaction ratings:

As Rudds uncommitteds have been crystallising out, they’ve been roughly breaking even since February, moving in equal proportions to satisfied and dissatisfied with his performance. Nelson on the other hand had a big chunk of uncommitteds move in February straight into the dissatisfied column that has remained at a pretty consistent level since March. Since the end of March, the uncommitteds that are crystallising out have actually been moving to the “satisfied with Nelsons performance” column which is pretty interesting considering the beating he’s been getting of late.

The size of Nelsons uncommitted number is also large at over 1 in 4 voters.

So that’s where we are all currently at in the polling. Anyone have any theories or insights over how the data is playing out with the politics?

AddThis Social Bookmark Button

add to kwoff

Posted in Polling, Voting behaviour | Tagged: , , , , , | 17 Comments »

Political Advice By The Column Inch

Posted by Possum Comitatus on April 1, 2008

crikeylogo.jpg This was me earlier today in Crikey.

In a headline that rivals “Strange Man on Public Transport!” for its sheer obviousness, Opposition is a Tough Business. With new governments come new oppositions which generally struggle to cope with the large decrease in relevance associated with the opposition benches. But after eighteen weeks, most Federal oppositions have at least developed some veneer of political strategy, some understanding of the job required in opposition which the polling starts to reflect. The day to day demand of having a 5 second grab on the great suite of topics that make up the news cycle starts getting complimented with more strategic approaches to the long term business of opposition.

What seems to separate the current opposition from their forebears is that the political strategy in its entirety appears to have been outsourced by the column inch to a set of News Limited journos that give Hawker Britton a run for their money in terms of pure spin. We’ve had the carers payment “crisis” which was little more than journalistic speculation gone feral, we’ve had the Aurukun/Macklin nonsense, we’ve currently got the Australia/Japan relationship “crisis” where the list goes on and on and on. The problem is that these stories sit somewhere between manufactured outrage and mocumentraries on the quality spectrum, allowing the government to easily adapt to whatever crisis they’re apparently facing this week by throwing some small bone to kill the story – an early budget clarification on the one hand, organise a quick Japan meet and greet on the other.

While it’s to be expected that oppositions follow the news cycle, and its to be expected that this type of sensationalist tabloid journalism that drives eyeballs to advertisers will make up a large part of that news cycle, regardless of the size of the paper the stories are printed on – the problem for the opposition is that it’s mostly vacuous fluff that that the public either sees through, doesn’t care about, or worse – they do believe it was an issue and then watch as that nice man Mr Rudd far from caving in to pressure, simply does what’s right and ends up looking in touch with the voters.

If we create a rolling two pollster average using Newspoll and Morgan and compare the first eighteen weeks of the Rudd and Howard governments, something stands out:


By this time in the term of the Howard government, the Beazley opposition had started to move on from the easy pickings of the news cycle and began to compliment that by applying greater strategic pressure about the new government’s policy program, which resulted in Howard’s polling honeymoon being slowly eroded. Yet the current opposition with its scatter gun style and lazy strategic approach is, if anything, falling further behind the ALP as time goes on.

If we want to place it in an even starker context, we can compare the vote gap that existed between the government and opposition of the day in 1996 and 2008 – again using this rolling two pollster average.


Whether this is the result of Rudd being a better political manager than Howard, Beazley being a better opposition leader than Nelson, the nature of political circumstance at the time or some mix of any and all of these things – what is inescapable is that Nelson is failing and that’s not good for the quality of governance.

What might be worth a shot is for the opposition to spend a little more time focusing on real policy issues that the public actually gives a hoot about and a little less time following the droning choir of News Ltd spruikers that are taking tabloid politics to whole new shallows of gravitas.

Unless of course the Libs really like turning the previously unheard of 20 point vote gap into a regular theme of federal politics. They should look north and see how that’s played out in Qld to disabuse themselves of any notion that such a thing would be impossible.


In other news – Steve Dickson, one of the 8 State Parliamentary members of the Qld Liberal party has threatened to quit over the proposed party merger not being taken to the vote in the party membership. The good news is that such a move would break the 4 all deadlock over the regular Lib leadership tussles, avoiding the need for future coin tossing to solve this most difficult of issues.

Still on Qld matters, Lawrence Springborg has threatened to take his pineapple and go home if the Libs and the Nats continue to refuse to take his proposed new party seriously. Meanwhile Mal Brough has decided to storm the barricades of the Liberal organisation in QLD and put a sword to the evil forces of Count Santo Santoro and his dark army of mediocrity.

Not to be out done in the loony-tunes stakes, NSW Liberal State MP Ray Williams has been accused of getting all hairy chested and challenging a branch president (as well as anyone said president could muster for help) to an old fashioned round of fisticuffs. And just in case you thought that this outbreak of the sillyseason was limited to State politics, the NSW Libs at the local council level have started recruiting One Nation hacks to help them run their local government campaign in the Baulkham Hills area.

Meanwhile, away from Tory central and over at the comrades in Victoria, Andrew Landeryou has uncovered some nuttery going on in Higgins at the local ALP branch level that pretty much explains why most people couldn’t be buggered to join political parties.

And finally, Andrew Bolt plays an April fools day prank on his readership, demonstrating via the comments section what most of us have known about his particular audience for a long time.

On something completely different – this is why they made Youtube.

Ooops – sorry. That was just Nightwatchman threatening to get “in and out of everyday Australians”. If you’re heading to a servo or shopping center over the next week or two, try not to get violated.

This is why they made Youtube!

AddThis Social Bookmark Button

add to kwoff

Posted in Crikey, Polling | Tagged: , , , , , , | 15 Comments »

Chilling Steadiness and Deadly Intent

Posted by Possum Comitatus on November 20, 2007

Don’t you just love that cracker of a line from Peter Hartcher about the behaviour of the polls?

As a new day dawns, the institution of Newspoll Tuesday – campaign edition, comes to a close.

And what new excitement does the Newspoll bring us on this solemn occasion?

None at all – its more of the type of business as usual that Peter Hartcher described so well.

The headline figure gives us a TPP estimate of 54/46, with the ALP down 1 and Coalition up 1 since the last Newspoll.


On the primaries we have the ALP down 2 to 46, and the Coalition up 1 to 41, giving us the following (thanks to George P. again). These results come from a survey of 1696 people giving us an MoE of around the 2.5% mark.


The Greens vote is at their highest level for a long time, at 7 points matching their 2004 election result of 7.2. The non-green minors are stuck on the same 6 points they have been for the last 4 polls.

So Newspoll Tuesday – ‘campaign edition’ ends with the same more of the same that it’s been delivering for weeks.

But the one thing I wanted to show you all today was the long term trend since the 2004 election. If we take the ALP two party preferred vote between November 2004 and November 2006 and run a linear trend using a simple regression through that period, and then we project that pre-Rudd trend through to today – look at what we find:


The ALP two party preferred vote is currently at its long term, pre-Rudd trend.

Now isn’t that interesting.

AddThis Social Bookmark Button

add to kwoff

Posted in Polling, Voting behaviour | Tagged: , | 25 Comments »

The Polls Have It

Posted by Possum Comitatus on November 16, 2007

Two bits of polling to chew through today, an ACN phone poll and a cumulative Newspoll breakdown.

First up we have a new ACNielsen, with the ALP primary down 1 to 47%, the Coalition primary up 2 to 43%, with the minors and others down 1 to 10%.

The headline two-party preferred results have the Coalition up 1 point to 46 for a 54/46 TPP lead to the ALP.

Not much going on there in terms of movement, with 1465 survey respondents giving an MoE of about 2.6%.

Tracking the primary votes over the longer haul we get this (Ta George):


And a TPP history of this:


ACN since August has had 55/45 being the name of the game, coming down from a 57/43 split in the first half of the year.

There’s not much to look at there, so moving right along to the Newspoll breakdown over at The Oz which was taken either side of the rate rise, but before the campaign launches.

The Newspoll cumulative is based on the last two big Newspolls to give a combined sample of 3402 respondents, giving a MoE on the state breakdowns at about 4%. Here I’ll take the first two Newspolls and combine them into one result, then take the last two Newspolls and do the same to get two different polls that don’t overlap (Newspoll is running a two poll rolling average, but I’d rather distinct pieces of polling data here).

Tracking the primary votes over time we get (where week 2 and week 4 refer to the campaign):



Interesting here is that the WA primary has been on an increasing trend for both parties over the last few months at the expense of the minors vote. The other states are showing a fair bit of volatility with that 4% MoE making it hard to tell what’s really going on if anything. But what is probably certain is that the Primary vote for the ALP in NSW didn’t drop by 5% in 3 weeks (nor increase in Vic by 5% over 3 weeks) – that sort of movement just doesn’t happen that dramatically without something like a leadership change, so there’s a fair bit of noise in the series.

Looking at how this transfers across to the TPP:


This would deliver 99 seats to the ALP in a new parliament according to Antony’s spiffy calculator, which is 9 seats more than the number that a national 6.8% swing would deliver using just the national pendulum.

Qld delivers 13, NSW 7, Vic 9, SA 5, WA 2, Tassie 2 and the NT 1.

Higgins would fall in such a uniform state swing and Goldstein would become the most marginal electorate in the country, being retained by the Coalition by 0.03%.

That would be Howard, Turnbull, Costello and Brough all sacked by the electorate leaving an interesting decision for the Libs in terms of who would be leader.

It’s interesting to see the non-capital city blow out from the capital city swings over the last few months – that’s probably an artifact of the Qld swing on the one hand with Qld being rich in regional seats, and the 35-49 females in NSW city seats on the other, which we better explain with the demographic data.



What stands out here is the movement in the female vote back to the Coalition.

Taking this into account with the state swings, we can probably say that Newspoll was picking up a change in 35-49 year old female voters primarily living in NSW city seats (If I were to make a guess). If there is a bit of sampling error in the polls to explain the big jump in NSW, this looks to be the demographic where it occurred. The 35-49 group has moved 7% back to the Coalition over the last few months, with females moving 5% over the period (males only 1%) and NSW moving 5% over the period as well – all back to the Coalition primary. When you take account of the capital city vs non capital city swings from earlier, it seems to be a metropolitan phenomenon.

In terms of primary vote swing since the last election, as close as we can tell, they look like this:

  Male Female 18-34 35-49 50+
Coalition Primary Swing -4% -4% -6% -1% -6%
ALP Primary Swing 8% 8% 12% 5% 8%

So you can get a bit of an idea of the demographics where the minor parties are taking a hit.

If this female NSW swing is happening (which it may be, just not as much as the data is probably suggesting) it will be interesting to see if the bank notices arriving about increasing mortgage payments makes an impact (especially as surveys suggest it is the female in the family that does the home budget), as well as how the childcare and education policies from each party play out.

As for what’s going on in Victoria – your guess is as good as mine. Those Mexicans have been confusing all year and don’t look like stopping any time soon.

AddThis Social Bookmark Button

add to kwoff

Posted in Polling, Voting behaviour | Tagged: , , | 40 Comments »

Why it’s all about John

Posted by Possum Comitatus on November 6, 2007


This was me in Crikey yesterday.

Since February, the Coalition political strategy has played out on the ground as an attempt to focus attention on Rudd. Whether this has been more by accident than design is probably worth pondering as well, but for all the “look at Kevin” programs, not a great lot has been achieved.

From Rudd dining with Brian Burke , his childhood memories, his links to those union blokes that keep turning out the lights, right through to actual policy programs like education, environment and infrastructure initiatives (that we now know, courtesy of the infamous Crosby Textor Oztrack 33, actually worked in the Labor’s favour by highlighting issues that the ALP had position dominance on) – the strategy that actually played out on the ground was one of focusing attention on Rudd.

Yet for all those attempts at focus shifting, and for all electoral diversional therapy involved, the key measures that matter continue to be intimately linked to the performance of John Howard himself.

The Coalition two party preferred result continues to be intimately linked to Howard’s satisfaction rating since the 2004 election.


A few quick regression equations and a granger causality test on the relationship between these two series suggests that it is the change in PM satisfaction levels that leads to changes in the Coalition two party preferred vote, rather than the two series moving together as a result of third party influences.

The punditry may say that Howard is still extremely popular considering his satisfaction ratings, but with his satisfaction ratings being so intimately linked to the Coalitions TPP vote, that line of thinking quickly becomes a bit a grand non-sequitur in the general scheme of things. His satisfaction only needs to fall small amounts to have a serious impact on the Coalition vote.

The other key measure intimately linked to Howard’s performance is the ALP primary vote via the PM dissatisfaction rating, as we hinted at last week.


Again, after a few quick regression equations and a granger causality test on the relationship between these two series, it is the change in PM dissatisfaction levels that leads to changes in Labor’s primary vote, rather than the two series moving together as a result of third party influences.

What is also interesting to note here is the gap that has recently opened up between these two measures, suggesting the possibility that Rudd is starting to gain support as a result of what the ALP is actually doing, rather than simply relying on dissatisfaction with Howard to deliver them electoral support.

So while all the policy noise and political advertising fills the political brainspace of the nation, when it comes right down to it, this election is still all about John Howard.

The big danger however is that hint in the last chart that suggests that Labor might finally be gaining support on the basis of their own merit. If that relationship starts to consolidate, there will be very little that Howard can do to turn his electoral fortunes around. When you are staring down the barrel of electoral annihilation, that is probably the last thing he wants to hear.

AddThis Social Bookmark Button

Posted in Crikey, Leading Indicators, Polling, Voting behaviour | Tagged: , , , | 4 Comments »

Bring Out Your Dead

Posted by Possum Comitatus on November 1, 2007

Today we have the quarterly Newspoll you have when you aren’t having a quarterly Newspoll – a sample of 3413 voters taken over the last fortnight, broken down into States, city type and demographics.

The State breakdowns have an MoE of around 4%, the age breakdown around 3% and the capital city vs non-capital city 2 and a bit %.

So let’s take the longer view first, looking at the behaviour of the States since the lead up to the last election, and where the entry for the 4th Quarter of 2007 is todays Newspoll results.



Since the last quarterly breakdown we’ve seen Victoria come back to the fold a bit for the Coalition, Qld moving solidly to the ALP and WA moving a tad to the ALP. But considering the MoE on these things, it looks like we have 4 states bunched up around an ALP primary of near 40 50 (bloody hell, how’d I do that), and WA down around the very low 40s.

The Coalition though is stuck down around the high 30’s and low 40s everywhere except WA. No election is going to be won with those low primaries.

Next up we’ll look at the demographics over time:


Slight movement to the both parties on the primary vote here is the name of the game, with the 18-34 demographic being the biggest mover, with a 4 point jump to the ALP. Maybe noise, maybe not.

Now for the swings.

First up, primary vote swings by State:


Qld – Yikes!

Not that NSW or SA are particularly pleasant for the Coalition either. I reckon the Qld number is a bit too big and the Vic number a bit too small considering what I’m hearing on the ground in both places.

On the demographic front we have:


There’s two big movements here of interest. Firstly the 6 point decrease in the minor party vote that’s gone to the ALP in the 35-49 age group. The other is the 8 point decrease in the minor party vote in the non-capital cities that has flowed to the ALP primary. If that holds, that would have to be a lot of the hippy green and rural/regional independent vote moving across, net, to the ALP.

And finally the big one – the TPP swing by State:


If this were repeated on election night, it would lead to around a 42 seat gain to the ALP and 102 ALP members in Parliament.

Qld alone would, should these state swings be uniform, deliver the ALP government with 17 seats falling up here. NSW is next on 14, then SA on 5, WA on 2, Tassie on 2, NT on 1 and Vic on 1.

Elsewhere the exceptional Mr Meganomics has a great article on some Newspoll data in terms of actual numbers of people over HERE


You can see the results of this in terms of which seats would go by visiting Antony Greens spiffy election calculator, with the results pre-linked in HERE

AddThis Social Bookmark Button

add to kwoff

Posted in Polling, Voting behaviour | Tagged: , , , , | 81 Comments »

Newspoll MkII

Posted by Possum Comitatus on October 30, 2007

An interesting little question popped up in Newspoll, and to fully chew it over properly, it needs to be put in the appropriate context:

ratesandvotes1.jpg ratevalpprim1.jpg

It looks like most people are pretty locked in on this question, with even numbers of ALP and Coalition supporters suggesting that they may cross sides in the event of a rate rise. The alternative is that maybe people don’t realise how they will act until they actually see the money coming out of their bank account.

History gives us a good suggestion here of how people actually react to rate rises as opposed to whatever they may say about how they would react.

The other thing of note on the Newspoll vote was the low minor party support. This gets us on to how the small minor party vote estimates are highly volatile because they are so small.

To highlight this, we’ll go through a two step process. First, if we subtract the TPP vote of Coalition from the TPP vote of the ALP for every Newspoll since 2006, we’ll get a TPP spread. Then if we do the same for the primary votes, we’ll get a primary vote spread. These two together show us how much of the difference in the TPP vote is explained by primary votes. If we then subtract the primary vote spread from the TPP vote spread, we end up with a figure that represents how much of the TPP spread is caused by the minor party vote.

That might sound a little complicated, but it’s pretty easy once you chart it.


If we blow that minor party residual effect up to highlight its movement we get:


What this chart tells us is how much of the TPP spread is explained by the minor party vote. This minor party vote is relatively volatile as a result of it being a small number in a larger pool. For instance, if there was a survey of 100 voters and 40 were voting for party A, 40 for party B and 10 for party C, if in the next survey it turned out that an extra person was voting for party B, the vote for party P would become 41, which is a 2.5% increase, but a 1 person increase in Party C (from 10 to 11) would be a 10% increase in the Vote of Party C.

With the sizes of the polling samples we use in Australia, this effect never fully washes out of the system. So small party votes estimates are, proportionally to their vote size, more affected by sampling noise than the major party votes.

As a result of that, and because TPP estimates are based on the distribution of those noisy vote levels, a fair bit of the TPP spread is quite noisy. Hence the TPP estimates bounce around more than they would be moving in reality.

AddThis Social Bookmark Button

add to kwoff

Posted in Polling, Voting behaviour | Tagged: , , | 34 Comments »