Possums Pollytics

Politics, elections and piffle plinking

Leader Dissatisfaction and the Minor Party/Undecided Vote.

Posted by Possum Comitatus on June 22, 2007


This caught my eye the other day. The red line is the net leader dissatisfaction value given by the PM dissatisfaction level minus the Opposition leaders dissatisfaction level as determined by Newspoll. The black line is the minor party primary vote and the undecideds also given by Newspoll.

Between 1996 and the Ryan by-election, the level of the minor party votes and undecideds combined tracked the net leader dissatisfaction levels extremely well.The Ryan by-election started a divergence and by February 2002 the two series went their separate ways. As can be seen by the below graph which shows the dissatisfaction ratings of the two major parties, while the PM dissatisfaction rating roughly tracked the minor party vote, the difference between the two dissatisfaction levels really tracked the minor party and undecided votes over the 1996-2001 period.


I’m not quite sure what this means, but if anyone has a good explanation of what dynamic could have been at play here – I’d love to hear it.

Another piece of related data is the total dissatisfaction levels compared to the minor party vote + undecideds, where the total dissatisfaction levels are simply the dissatisfaction ratings for the PM and the Opposition Leader added together.This is graphed below:


This seems to show that while the satisfaction levels with the two major parties does in fact drag the minor party and undecided vote level around; it doesn’t actually do it as much as I was expecting to be the case. There are a number of spikes (both up and down) in the dissatisfaction levels that weren’t matched by movements in the minor party + undecided vote level. The electorate seems to regularly say to the major parties “I think you’re shit, but you still get my vote”.


Economan thoughtfully pointed out that I had indeed missed the obvious downward spikes in Total Dissatisfaction levels (dont ask me how I did that) and that this could well mean that the voters are saying to the major parties “I think you’re OK, but I’m going to tell the pollster I’m voting Green/Dem/whatever”



7 Responses to “Leader Dissatisfaction and the Minor Party/Undecided Vote.”

  1. EconoMan said

    On the last graph, the large spikes are all on the low side, rather than above the black line. Rather than “I think you’re shit, but you still get my vote” that seems to suggest (at least at those times):

    “I think you’re OK, but I’m going to tell the poster I’m voting Green/Dem/whatever”

    Given total dissatisfaction averages around 100-110 — let’s say 55 dissatisfied with both parties — that crudely leaves around 10-20% dissatisfied with both parties. About the mean minors + undecideds?

  2. Possum Comitatus said

    Spot on about the downward spikes – for some reason I was obsessed with the upward spikes at the time.I’ll update that to point out the obvious that I so clearly missed.

    On the last bit – sounds like a graph!

    But I seem to be having a blonde moment – when you say “that crudely leaves around 10-20% dissatisfied with both parties”, I’m not getting you.Any chance you could explain what you’re getting at a bit further?

    Then I’ll whack it together as a graph and stick it up.

  3. EconoMan said

    Well if 55% is dissatisfied with major party 1, and 55% are dissatisfied with major party 2, that gives you a total dissatisfaction of 110 (duh).

    A little mental ven diagramming might suggest that 40% might be dissatisfied with MP1, but satisfied with MP2. And 40% the reverse. That leaves 15% in the overlapping part: dissatisfied with both.

    Mathematically: DissMP1 = 55, DissMP2 = 55 DissMP1 INTERSECT DissMP2 = 15.

    For completeness, that would leave 5% worth of ‘undecided’ UNION ‘satisfied with both’.

    It would be the 15% dissatisfied with both that are most likely to vote minor party or be undecided.

  4. Possum Comitatus said


    Unfortunately Newspoll doesn’t contain data rich enough to pull that out – which is a pity because it could easily be done.It would be nice if the Newspoll results contained individual IDs on the responses.I also wish that Newspoll would ask people that have changed their voting intention recently where they were planning to vote before.


  5. EconoMan said

    Yeah, I know you can’t actually do that. I’m just guessing /hypothesising that what I said is about right.

  6. Possum Comitatus said

    Well your guess sounds as good as any and better than most!

    The problem with any of the satisfaction dynamics is they’re polluted by the 75% thereabouts of people who are rusted on to their voting behaviour in practice, even if not in principle.

    It can take only small changes in the satisfaction levels to have major effects on primary voting intentions, but only if those people that have changed their satisfaction ratings are also swinging voters.Likewise, large changes in satisfaction ratings can have a meaningless influence on primary voting intentions if the people changing their mind on satisfaction ratings are rusted on voters to one of the parties.

    And looking back over 21 and a half years of Newspoll data, rusted on Coalition supporters saying, for example, that the ALP leadership is doing a good job seems to be quite a frequent occurrence, meanwhile the ALP vote doesnt move because the people saying they have high satisfaction ratings arent the ones that count (as far as the ALP is concerned).That was one of Beazleys great problems.

    It also seems to happen with the Coalition and ALP supporters, but to a much smaller extent.

  7. Terima Kasih Atas Informasinya Salam kenal

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: