What The People Want: July 2007 Archives

« June 2007 | Main | August 2007 »

31 July 2007

Lapsed Liberals - the ebbing tide

This is analysis of those in our sample who have changed their intention since last election from Liberal or National to preferencing Labor.

The Coalition’s support has literally ebbed away. While you might expect that there would be voters changing in both directions - from Coalition to Labor and vice versa - this election the traffic is just about all one way. The Liberals' vote is as low as it is because people who voted for Howard last time have changed their minds and gone across to Rudd. Why has this happened?

There are a number of clues in our responses.

First, there is no obvious demographic group from which they come. When you look at occupations, the splits are very similar to our total sample. Teachers and the retired are still the two largest groups, and every other group is represented within the tolerance of sampling error. I’ve scanned postcodes, and there is nothing which suggests a bias towards outer-urban or rural electorates, and representation from the states is well within sampling bias. Likewise with income. This group is very typical.

Howard’s grip has loosened across the whole range. Previous election victories have been defined by capturing key demographics: like blue-collar conservatives, the elderly and young families, using combinations of special benefits - baby-bonuses and retirement benefits. Or by touching deep cultural nerves - the “elites” versus the “real” Australians. Now representatives of most of these constituencies are defecting.

The only quantitative difference between this group and the whole sample is in the gender split - males appear to be over-represented, particularly between 35 and 64 years of age, as well as in the under 24s. I’m not sure what this reflects, but the government has been running an anti-domestic violence campaign for months, so perhaps it reflects the success of those ads with younger women.

Second, these people are on balance pessimistic about the direction of the country, mildly unfavourable to Howard, and appear very enthusiastic about Rudd. While 71 per cent approve of Rudd, only 30 per cent approve of Howard, and 59 per cent think the country is heading in the wrong direction. Some days ago John Howard asked his cabinet if the problem was “him”. It appears that it is. Or at least, when these voters hear mention of the Howard Government they compare it unfavourably with the prospects of a Rudd government.

Third, these voters are more likely to be Christian than our sample in general. This suggests a less materialistic mind-set than that of voters in general. And while they are not particularly evangelical in their attitudes, on balance (52 per cent to 40 per cent) they believe that Christian values have a place in political debate. They also overwhelmingly believe (77 per cent of them) that Christian values have been valuable to society.

So while they don’t appear to have a particular geographical home, they do appear to share some deep attachments to Western conservative values. This is reinforced by the fact that only 9 per cent of them claim to be traditional Labor voters. Over half identify as traditionally Liberal or National, and 22 per cent as having no traditional political commitment. This possibly explains the seemingly inexplicable combination of lack of anger against the government in the electorate and the large swing to Rudd. Many who make up the Rudd majority are voting against their normal political family.

This makes it a soft vote. Even though everyone in this sample says they are not voting for the government, 15 per cent still wants the Coalition to win, and only 57 per cent wants Labor to win. Perhaps there is some hand-waving going on here. The statistics suggest that these voters haven’t so much deserted the Liberals, as that they’ve crossed to the other side of a valley and they’re trying to encourage the Liberals to follow after them, or give them a reason to come back.

The qual fleshes out some of these perceptions. Major concerns about the direction of the country centre around a sense that society is increasingly divided between the haves and have-nots and a concern for the environment, particularly water and climate change. The government is seen as aligned with business and the rich, and uncaring about the poor. IR is present as a supporting issue to this perception. There is also concern about our foreign policy, and it tends to be expressed as part of an overall concern that we are becoming too American in our society.

While respondents are concerned that the economic benefits should be spread around there are no suggestions as to how this might be done. Economic good times are being taken for granted, and they are providing many Australians with the opportunity to be post-materialistic in their approach to politics. This makes the government seem old-fashioned. In focusing heavily on the economy it appears to be located in the industrial-age rather than the modern world.

These concerns are reflected in the issues that respondents emphasise as being important. At the same time there are some interesting conflicts. Climate change is an issue that cuts in both directions. While many blame the government for not acting decisively, others blame the government for acting at all. Why the latter group is veering towards Labor is not at all clear if this is the important issue for them. Climate change is also an issue that emphasises the perception that the government is focused on the past rather than the future and has failed to plan.

While the ACTU has been running a concerted campaign on IR laws, it is only when asked what would make them hesitate about voting for Howard that these lapsed liberals put IR at the top of the list; and then it is a very long list where many issues get almost as many mentions. This suggests that on its own it is not a major concern, but if they have to provide a justification for their voting intention, then it becomes more prominent. Yet even then, a word search shows that IR ties with Iraq for the number of mentions, and both of them are very much a minority concern.

The major theme that runs through these swinging voters’ hesitations in voting for Howard is a perception that he is arrogant and has lost touch with “Strugglesville”. For each of them it seems to be embodied in a slightly different issue. Again, this is a soft motivation, because when they are asked what makes them hesitate in thinking of voting for Rudd, arrogance is a strong theme. This is underlined by his “mini-me” strategy where voters see him as being just a Howard clone.

Even stronger as a hesitation is Rudd’s lack of experience and a feeling that he does not have the mental toughness to do the job and stand up to his colleagues. As you would expect with a group that normally identifies as Liberal, there is a fair degree of distrust of the Labor Party, another major negative for Rudd.

Analysis of this group suggests that the election is not yet won by Rudd. A tide which ebbs in one direction will just as easily flow in the other in a few hours time. For Labor to keep hold of them it needs to address a multitude of issues, many of which run counter to others.

By contrast, the negatives about Rudd are all focused in a few areas. It would be relatively easy for the Liberals to craft a few messages that emphasised those negatives and turned the tide, and relatively difficult for Labor to pull them back. If all they are offering is a younger clone of John Howard, then it’s possible voters will decide to stick with the original. After all, they already know its defects.

In fact, if they vote for Howard this election he will be gone in another two years, and then they can get a really clear focus on the future and make a decision between two future leaders who will have had the chance to prove themselves. It’s possible that this election may end up being an audition for the next.

Posted by Graham at 05:27 PM | Comments (6)

24 July 2007

Note on the Polling Wars

Recently The Australian ran a full editorial criticising the treatment of its polls by some online commentators (I suspect not including us, given that I was quoted in the following day's edition on another matter).

The debate was largely one of "sound and fury" and centred on interpretation of Newspoll data by The Australian's Dennis Shanahan. Shannahan claimed that a recovery in the Liberal Party vote in federal elections was always pressaged by a recovery in Howard's personal vote. This was disputed by a number of bloggers, including political scientist Peter Brent, who directly, or indirectly, fed into Crikey!. The counter argument was that Howard's approval rating rose as a consequence of an increase in the Liberal vote, not as a causative factor, or that there was actually no relationship between the two.

In fact the debate is irrelevant. No two elections are the same, and it depends how they are run as to whether the leaders' ratings matter, and what effect they have on the party vote, or it has on them. In the 1995 Queensland state election the Coalition went in with approval ratings for Rob Borbidge of 19%. So we didn't run a head-to-head leadership campaign because that would have meant losing to Goss who was at the time Australia's most popular politician.

It would be smart for the coalition not to run a head-to-head leadership campaign this time either. At the moment voters favour Rudd, but they're worried about his Labor Party and union connections, which is where the government should direct its campaign, in which case, Howard's personal approval rating is of much less significance.

All of which raises the question of how journalists and others read polls. At the moment a number are writing up the proposition that Howard has gone bad with the under 34 year olds. He may have, but the Newspoll figures (on which these stories are largely based) are equivocal at best.

Date	18-34	35-49	50+
Election 04	40	42	52
Now	30	35	44
Swing	10	7	8

This table suggests that the swing is greatest in the younger groups, but sampling error negates this. Newspoll advises that the sample size for those between 18 and 34 is somewhere around 250, meaning that the sampling error will be somewhere in the vicinity of 6.5%! In effect, the swing in all of the groups is statistically the same.

Perhaps journalists have been misled by the 1.5% sampling error cited in the footer of the poll. This figure refers to the error of the overall sample. The sample consists of 6,957 respondents, but they are spread over 6 separate polls, and something like 12 separate weekends. As a result each figure has a much higher sampling error than the whole sample, and it is in fact difficult, if not impossible, to think of any statistics that can be extracted from the tables where 1.5% would actually be the sampling error

Posted by Graham at 11:05 AM | Comments (0)

23 July 2007

Why do so many teachers fill-in our surveys?

I note in my report that we have a high percentage of respondents who are teachers. One of our teacher respondents has some ideas as to why this is the case. Is he on the money?

As a teacher / Deputy Principal perhaps our over representation is due to:

The need to be somewhat tech / Internet savvy as this is increasingly incorporated into curriculum and this is the world that young people dwell in.

The understanding of political action / inaction on the future of society. Teachers are on the coal face of preparing the next generation for the world – we have some insight as to what the future might look like. We are in essence sick of increasingly crowded curriculum and know the implications of politicians who endeavour to fix yet another societal ill through the school system (speeding, drug taking, illicit sex, violence, plagiarism …) without any further funding, without an increase in time and where the benefit is potentially more to the politician who makes the decision than to the teachers, students and parents who are affected by the decision. We also get a snapshot into what families look like and the problems that beset them. We understand a little of the issues that will be facing the next generation of adults as we work with them now and see the signs and symptoms of what is to come – we actually have insights into what might be solutions to this problem. We are sick of politicians who don’t appear to listen and hence when someone offers to listen to our opinion we are willing to share to. We are not mere bystanders in this process – we are in education to make a difference. Teachers are generally people of reasonable initiative and creativity and are focussed on making an all-round tangible difference. We believe in finding real solutions to real problems. We have a long term view of the solutions required which must stretch beyond a 3 year political term and also we understand that the court of public opinion is not always the best place to develop policy (eg – performance pay for teachers).

Sadly education debate in NSW continues to be an ‘us and them’ scenario between school sectors perpetuated by ‘lies, damn lies and statistics’. Schooling is under funded in a state where education costs are the highest on average. The AEU in particular continues to politicise education in a way that ultimately does a disservice to both sectors. They are both struggling and need some long term vision and funding which the cross sector bickering fails to provide.

Posted by Graham at 02:34 PM | Comments (6)

11 July 2007

Religion by age

In response to a poster using our data to determine the age profile of the various denominations I used the current census data to construct this graph. Sorry about the width, but I couldn't see how else to have every denomination labelled. You'll need to click the comments button below to see it in all of its glory.

Posted by Graham at 11:39 PM | Comments (8)

9 July 2007

Federal Benchmark - the religious vote

This is the second instalment of analysis of our Federal Benchmark survey and deals with the Christian vote. You can download a copy of the report by clicking here (PDF 150kb).

Christian respondents demonstrate very similar demographic characteristics in age, gender, occupation and income to the total sample. However, they are more likely to vote for the Coalition across the sample, with Coalition support being strongest amongst the evangelical protestant churches. Catholics as a group are strong Labor supporters.

This sample picks up a larger percentage of evangelicals than are present in the general population.

While it is impossible to generalize across the entire sample of Christians on most issues, it does appear to be true, no matter how they vote, that they are concerned about issues of morality, and care and compassion. The reasons that they give for their voting intentions reflect this, even when they are voting in opposite directions. It also seems to be the case that the more frequently someone attends church, the more likely they are to give their vote to an independent party, rather than a mainstream one.

In general they also appear to be more concerned about the Labor Party because of its policies than the general public. However, they are very attracted to Kevin Rudd and the evangelical segment is also drawn to candidates of similar beliefs. Rudd’s Christianity works in his favour, but Howard’s appears not to work in his.

Key findings are:

Christian respondents share similar demographic characteristics to the total sample, including such markers as the over-representation of teachers as respondents.
Anglicans and Catholics are under-represented in the survey, with Pentecostals being significantly over-represented.
The distribution of Catholics by age is similar to the total sample, Anglicans and Uniting Church are older than average, and the evangelicals are younger than average.
Anglicans and Catholics are less likely to be regular attenders, and Uniting church the least regular. By contrast 60% of Pentecostals claim to attend more than once per week.
Catholics are strongly Labor (50%), Anglicans tend more towards the Coalition (32% Labor and 32% Coalition), Pentecostals are the home of Family First (36%), and Labor does poorly with the rest, although this does not reflect directly in an increase in the Liberal vote, but rather flows to parties like Family First and the Christian Democrats.
Catholics appear to be swinging against the government more strongly than the other denominations, apart from Baptists. Uniting are also above the norm, and Pentecostals the least likely to swing.
Howard is viewed favourably by Christians in general, although narrowly (51% approve, 41% disapprove). There are significant differences between denominations. Evangelicals are strongly approving, Anglicans almost evenly divided, Uniting Church strongly disapproving and Catholics very strongly disapproving.
Rudd is viewed favourably by Christians in general. The highest disapproval figure is 31% amongst Prebysterians, and the lowest 16% amongst Catholics.
The desire as to who should win the election runs in line with voting intentions.
Expectations also are for Labor to win, apart from amongst Pentecostals.
While our complete sample believes that religious values have no place in political debate, a majority of Christians disagree substantially (64%). However 48% of Catholics do think religious values have no place in political debate, and only 42% think they do.
Most Christians believe that Christian values have had a positive impact on society (73%). Catholics are least likely to agree (64%) and Pentecostals the most likely (100%).
Most people, Christian or otherwise, are neither more nor less likely to vote for someone from their faith community (treating Agnostics and Atheists as a “faith”). However, 25% of Atheists and 22% of Christians say they are more likely to vote for a fellow-traveller, while 6% and 3% respectively say that they are less likely.
This generalization is not true of evangelical Christians where 39% (Presbyterians) to 90% (Pentecostals) were more likely to vote for someone of the same religious persuasion.
The tendency to support co-religionists increased with the frequency of church attendance.
Public profession of faith made the total sample less likely to vote for most politicians. Only Christians were positive with evangelicals being more positive.
Kevin Rudd suffers less from his profession of Christianity than other comparable politicians such as Howard, Vaile, Beazley, Costello, Boswell and Fielding.
The gap between Howard and Rudd is largest amongst Churches of Christ, Uniting Church, Anglicans and Catholics. Howard narrows this gap amongst Baptists, Presbyterians and Pentecostals. Atheists, agnostics and Buddhists are also more tolerant of Rudd’s Christianity than Howard’s.

Posted by Graham at 10:42 PM | Comments (21)

Federal Benchmark - General Analysis

I've finally uploaded the first of our reports on the benchmark survey which you can download by clicking here (PDF 326kb) I've also posted an article to On Line Opinion entitled "Death by Water".

Like the Fisher King, John Howard is in trouble because of the weather. Climate change and drought, seen by many as twins, are the most important issues to Australians. If they are not fixed then there will be no economy to boast about. This is a bigger risk than the possibility of interest rates rising. Labor has credibility on the climate. The Coalition generally has it on the economy. When the economy is viewed through the prism of climate it undermines the Coalition’s key strength. This helps to explain the sudden slump in their popularity independently of the accession of Kevin Rudd. It also suggests that Industrial Relations and the Unions Your Rights at Work campaign are not the major agents of change.

Incumbency is also a problem for Howard. After 10 years in power he has upset everyone at least once. They’re not actively seeking change, but they are grumpy.

Howard’s ascendancy has also depended on a deal with the electorate where in return for representing blue-collar conservative aspirations he has gained their support at the same time losing some middle-class supporters who are often snidely, and inaccurately, referred to as “doctor’s wives”. These “values” voters think Howard has reneged on the deal. He used to be one of them, but now they suspect he’s traded-in his chesty bonds and Y-front for some designer flimsies, and he’s no longer listening. What’s more the constant opportunism and sharp practices are making them surly. Howard’s idea of the future is the next election, and a good budget surplus. After 16 years of continuous economic expansion they’re looking for something grander.

The problem is that while Kevin Rudd seems much more likeable, and is promising some upgrades, they aren’t sure that he has the experience or the ability to deliver. He’s also tarnished by his traveling companions - Labor and the Unions. Perhaps this also is a reflection of incumbency. With Labor in power in all the states there’s not too many who haven’t been irritated by them at least once. While Labor is making much of the IR laws they aren’t a first order issue for most Australians. Certainly not enough to save the Labor brand.

It is difficult to tell from this sample whether the public has made up its mind to change the government, or whether they are stamping their foot in the hope that Howard will pay more attention to them. In 2001 they were desperately unhappy with Howard, but he changed direction on enough issues for them to forgive him between losing Ryan and winning Aston.

If he can rebuild some trust with values voters and reinforce doubts about Rudd’s experience, Labor and the Unions, Howard may still win; particularly if the drought breaks. Conversely, if Rudd can prove that he can control his allies and keep talking about the future, he may be able to hold his lead.

Major features of the study are:

Sample of 3,189 drawn from all over Australia with a significant over-representation of Queensland and under-representation of Victoria, but otherwise close to actual distributions.
Respondents are Internet savvy, with higher than average incomes, and more likely than the normal population to be teachers, retired or in white collar occupations.
If an election had been held amongst this group the two-party preferred result would have been 35% Coalition, 65% Labor, excluding those how have yet to make up their mind.
19% of those voting Liberal last election say they will vote Labor this election, with another 12% saying they are undecided. This represents an approximately 8% swing away from the government, broadly consistent with quantitative polls.
Only 25% of the sample agreed that the country was heading in the right direction. 63% thought it was heading in the wrong direction. However, 81% of Liberal voters and 66% of National voters thought it was heading in the right direction.
The two biggest issues shaping perceptions of whether the country was heading in the right direction or not were “Climate Change” and “Economy”. While the economy is traditionally a strength for the government, when seen through the prism of climate change it becomes a weakness, particularly as a significant amount of the climate change concern stems from water fears. No water, no economy.
Most respondents disapproved of Howard - 61%. Only 33% approved. Most respondents approved of Kevin Rudd – 61%. Only 17% disapproved.
The “balance of hate” favours Rudd. Only 42% of Liberal voters disapprove of him compared to 88% of Labor voters who disapprove of Howard.
54% of respondents think Labor will win, compared to 22% who think the Coalition will.
57% of respondents want Labor to win, compared to 30% who want the Coalition to win.
There is a clear preference for one party or the other to win, with only 7% wanting a hung parliament.
The most important issue for voters is the weather with most nominating water, climate change, or both. Other important issues are the economy and values.
Industrial relations is not rated as a leading issue.
Howard is dying the thousand cuts of incumbency. While IR does rate as a hesitation for voting for Howard, it is one of many issues, none of which is overwhelmingly important. After 10 years Howard has upset everyone at least once.
While voters approve of Rudd they don’t trust him to be able to deliver what he promises. They are also concerned about the Labor Party, unions and his parliamentary team.

Posted by Graham at 04:38 PM | Comments (16)