Pollytics

Politics, elections and piffle plinking

New Pollytrend Metric – Phone Pollster Trend

November 8, 2009 – 11:25 am, by Possum Comitatus

We all know about our Pollytrend chart – where we build a rolling average of the most recent poll from each pollster weighted by sample size, then run an aggressive locally weighted polynomial regression through the results to give us an adaptive trend line. While Pollytrend is good for picking up medium term shifts in voter sentiment, some of us question it’s accuracy when it comes to it’s actual level at any given point in time. As the two pollsters with the largest sample sizes (Essential and Morgan Face to Face) have a slight relative lean to the ALP in the vote estimates compared to the phone pollsters, it forces the entire Pollytrend line to have a relative lean as well.

Today we’ll solve that concern by introducing a new addition to our Pollytrend – the Phone Trend. The phone trend line is calculated a little differently to our all pollster Pollytrend line.

First and most obvious, it uses only phone pollster data.

Secondly, only the most recent two polls of each phone pollster get entered into the initial calculations. We use the most recent two polls because any given pollster’s “poll before last” still contains valuable information about the state of public opinion today, but the information is just not as valuable as their most recent poll.  In order to differentiate between “best” information (the most recent poll) and “second best information” (the poll before last) from any pollster – we simply take a rolling all phone pollster average of the two most recent polls from each phone pollster, but weight all results by time – so the further back in the past a poll was taken, the less weight that particular poll carries when we calculate the weighted average. For example, a poll taken 60 days ago will only have one tenth of the weight of a poll taken today.

Thirdly, every time a phone poll is released we calculate our new weighted average to build ourselves a time series.

Finally, as we do with the all pollster Pollytrend, we then run a Read More »

Comments (18) | Permalink

Morgan Adds Outlier Weight to Newspoll

November 6, 2009 – 5:18 pm, by Possum Comitatus

Morgan has cheekily come in a week early, giving us the first week of what is usually a two week Face to Face poll with the primaries running 51 (down 1) /32.5 (down 2) to Labor, washing out into a two party preferred of 61/39 the same way – a half a point increase to Labor. The Greens are on 9.5 (up 2) while the broad “Others” are sitting on 7 (up 1). This comes from a sample of 1050 giving us an MoE that maxes out around the 3% mark. This poll was taken over the period of October 31 to November 1 – the same period as the now infamous Newspoll that has caused so much hyperventilation by the usual suspects over the last week.

If we plug this Morgan result in with the other ultra-regular polls of Newspoll and Essential Report, we now come up to date as of November 1st.

ALPprimary lnpprimary

ALPtwoparty

Outliers are simply part of life for a pollster – they happen and there’s absolutely nothing they can really do about the odd one popping up. So the first thing to point out is that if the last Newspoll was an outlier – the probability of which has just increased – there is no Newspoll conspiracy here. We actually expect pollsters to give us a WTF moment every now and again.

What catches my eye about these three pollsters is the way the ALP primary vote most Read More »

Comments (58) | Permalink

Pollytrend and Odds and Ends

November 5, 2009 – 7:59 am, by Possum Comitatus

Throwing the last polling cycle’s worth of data into the Pollytrend algorithm, we can see that Newspoll produced no discernible change in the trend line – which is to be expected, since our trend here measures how all polls move together and it can adapt to outliers.

pollytrendnov

Also worth having a squiz at are three charts that I used in a Crikey article on Tuesday that shows how the three pollsters that have been in the field of late – Essential, Morgan and Newspoll -  have comparatively behaved, with a focus on the last 3 weeks.

alptwoparty

Read More »

Comments (16) | Permalink

Outliers Outliers – get’em while they’re hot!

November 3, 2009 – 6:43 am, by Possum Comitatus

Today’s Newspoll via The Oz comes in with the primaries tied at 41 a piece with the ALP down 7 and the Coalition up 7. This washes out into a two party preferred of 52/48 – a 7 point change from last Newspoll. The Greens are steady on 10 while the broad “Others” are unchanged on 8. This comes from a sample of 1149, giving us an MoE that maxes out around the 2.9% mark.

Is this an outlier?

Probably, and it comes from something Newspoll head honcho Martin O’Shannessy said to the Australian.

The majority of the change in the Labor primary vote is attributable to a fall in Labor’s primary among those aged over 50 under 50

The only issue over the last fortnight in media has been asylum seekers, and all the polling we’ve ever had on that issue suggests that it is the over 50’s that are the strongest supporters of tough border protection – and as yesterday’s Essential Report suggested, the over 50’s are also more likely to believe that the Coalition would do a better job at managing this issue..

If this wasn’t an outlier – we would expect the over 50’s to move strongest, so the composition of the poll is inconsistent with what we would expect to occur if the poll was, in fact, an accurate representation of the true state of public opinion.

There is also other evidence for it being an outlier. Firstly, we expect that, on average, 1 in 20 polls are out by an amount larger than the margin of error – which in Newspoll’s case is 3%. Further, we expect that, on average, 1 in 100 of all Newspoll’s will be about by a margin larger than 4%.

This is simply the probability statistics of random sampling in action.

Secondly, this issue has been Read More »

Comments (85) | Permalink

First big polling on Asylum Seeker issue

November 2, 2009 – 3:23 pm, by Possum Comitatus

This weeks Essential Report has the primaries running 49 (down 1) / 35 (down 1) to Labor, washing out into a two party preferred of 59/41 the same way – steady since last week. The Greens are on 8 (steady), while the broad “Others” are up 1 to 8. The missing point is due to rounding. This comes from a two week rolling sample of 2007, giving us an MoE that maxes out around the 2.2% mark.

Additional questions this week are on the asylum seeker issue and one on climate change. These come from a sample of 1122 for an MoE that maxes out around the 2.9% mark.

This is the long awaited polling on public perceptions of the asylum seeker issue and it’s politics – what is really worth noting is how the set of responses collectively paint a far more sophisticated picture of the country than a simple pro/anti asylum seeker stance leading into a pro/anti government policy stance and any consequent  pro/anti Labor/Coalition position.  Yet again, the public shows itself to be a little more complicated than some of our punditry muppets would have us believe. I’ve reordered the questions from as they appear in the actual report to start with the wider issues, before narrowing to the politics and finishing off with the ultimate political question on the issue.

The Federal Government is currently working with the Indonesian government to stop asylum seekers entering Australian waters. For each of the below statements that have been made about current incident of asylum seekers, please indicate your level of agreement.

asylum1

The cross-tabs tell us: Read More »

Comments (22) | Permalink

NSW State Polling

November 2, 2009 – 9:44 am, by Possum Comitatus

Newspoll over the weekend via The Oz released another rather hideous poll for the NSW government showing the primaries running 42 (up 1) / 30 (down 2) to the Coalition, washing out into a two party preferred of 55/45 the same way – a 1 point increase to the Opposition over the last 2 months. The Greens are on 12 (down 2), while the broad “Others” are on 16 (up 3). This comes from a sample of 1271, giving us an MoE that maxes out around the 2.8% mark.

Along with the voting intentions, this poll also had all the metrics running against the Rees government with his satisfaction down, dissatisfaction up and preferred premier dipping a point as well. On the other hand, Opposition leader Barry O’Farrell had a mixed bag with his satisfaction and dissatisfaction both up (at the expense of his undecideds), while his preferred premier metric bounced up 3 points to 36 – giving him a 5 point lead over Rees.

With the ALP primary vote so low and optional preferential voting running in NSW, the two party preferred continues to understate the size of the electoral flogging Labor would get were an election held and these primary vote figures repeated. It’s interesting that the O’Farrell just can’t  seem to pull his net satisfaction rate up very high considering the rather tragic nature of the NSW government.

The usual charts come in like this:

Read More »

Comments (6) | Permalink

Victoria State Polling

November 2, 2009 – 9:22 am, by Possum Comitatus

The Newspoll site (and presumably somewhere at The Oz) has a new Victorian State poll out with the primaries running 43 (steady) /35 (down 2) to Labor, washing out into a two party preferred of 57/43 the same way – a 1 point gain to the Labor party over the last two monts. The Greens are on 15 (up 3) while the broad “Others” are on 6 (down 2). This comes from a sample of 1151, giving is an MoE around the 2.9% mark.

In the usual charts below, you’ll notice that Ted Baillieu has a bit of a kick upward in his satisfaction rating, his dissatisfaction rating went down and his preferred premier level lifted – yet the Liberal Party vote contracted by 3 points over the period. Sometimes you can’t take a trick.

Not much to say about this except that it’s the sort of business as usual poll in Victoria we’ve been seeing for the last 6 months. The charts come in as:

Read More »

Comments (1) | Permalink

Net Access – More Than Just Smut and Piracy

October 29, 2009 – 10:15 am, by Possum Comitatus

With the debate about the National Broadband Network well underway, some of the dismissals of why a population would need higher broadband speeds (or as Kevin Rudd so quaintly called it at the Tasmanian Community Cabinet meeting, “Bandspeed”) – often bounce around somewhere between the ridiculous and the incredulous. It’s not only the economic and social value of bringing higher speeds to market at a decent price point that gets the sceptical treatment by those essentially not across the policy brief, but also scepticism aimed at the rhetoric the government has been deploying over the NBN that it is has the importance and consequences of a key essential service akin to roads, power and water, and should be treated as thus.

Of particular interest to this point is the idea that I’ve heard floated around on broadband subsidies for low income households being a possibility further down the track, and how that would fit into the wider policy mix that seems to be falling under the broad banner of the Education Revolution.

Yesterday the ABS released an interesting set of data on sporting, cultural and technological activity undertaken by 5 to 14 year olds over the past 12 months, broken down into a number of demographic cross-tabs.

Within this ABS release is some pretty interesting data on internet use by kids that’s worth looking at and keeping in one’s thought orbit when it comes to some of the issues surrounding the NBN.

First up, the basics – the proportion of 5-14 year olds that accessed the internet in the last 12 months, and broken down by urban geography.

Figures 1 & 2.

ageaccess geographyaccess

The big jump in net access for kids comes between the ages of 9 and 11. Also interesting is that total net access for the 5-14 year cohort starts falling off the further away you get from a capital city.

On the question of where it is that kids access the internet from, the results show some interesting trends across age. Read More »

Comments (28) | Permalink

Could the polls actually be undercooked for Labor?

October 27, 2009 – 11:38 am, by Possum Comitatus

Something interesting has been popping up in the polling with the two party preferred vote – there is a statistically significant difference between the two party preferred estimate when poll respondents get to allocate preferences compared to the two party preferred estimate when preferences are allocated on the same basis that they were at the 2007 election.

We have two pollsters that publish these two different ways of allocating preferences to get a two party preferred estimate – Morgan and Nielsen – and in both cases the same phenomenon occurs. There’s been 72 Morgan polls and 13 Nielsen polls since Rudd was elected, so we have a decent set of data to work with – certainly enough to imply that the polls might actually be a little better for Labor than the published headline results suggest.

UPDATE: Nielsen’s headline numbers use respondent preferences even though they measure both , while Morgan and Newspoll use 2007 election preference distributions.

First up, if we run the ALP two party preferred results of each pollster using the two metrics, we get:

roymorgan2 nielsen2

In both cases, the ALP two party vote is consistently higher when respondents get to allocate preferences than when we use the preference distribution of the 2007 election. It’s worth pointing out that the ‘headline number’ used when reporting polls is based on the 2007 preference distribution for Morgan and Newspoll, but not for Nielsen.

If we take the difference between the ALP two party results based on how voters say they will allocate preferences and the ALP two party vote based on the 2007 preference distribution for each pollster, it tells the basic story.

roymorgan3 nielsen3

If we now transform those raw results into the percentage of time that the voter allocated result is higher than the 2007 preference distribution result, and by how much we get: Read More »

Comments (28) | Permalink

Essential Report – Best to Manage Edition

October 27, 2009 – 6:16 am, by Possum Comitatus

This week’s Essential Report has the primaries running 50 (up 1) / 36 (down 1) to Labor, washing out into a two party preferred of 59/41 the same way – a 1 point gain to the ALP since the last Essential Report. The Greens are on 8 (up 1) while the broad others are steady on 7. This comes from a rolling two week sample of 1967, giving us an MoE that maxes out around the 2.2% mark.

Additional questions this week go to who is best perceived to measure various issues of national importance, as well as some questions on the CPRS. These additional questions come from a sample of 1097, giving us an MoE that maxes out around the 3% mark.

.

How important are the following issues in deciding how you would vote at a Federal election?

importance1

It’s worth noting how the importance of issues changes as those issues dominate the media cycle. Consider the issues that have been dominating the media cycle lately – things like border security, post-stimulus economic issues and the CPRS – and the categories they could broadly fit into have all lifted, with Security/War on Terrorism and Climate change lifting by large margins compared to public perception back in March of this year.

Leadership also had an interesting drop down the list over the period.

Read More »

Comments (5) | Permalink