Politics, elections and piffle plinking

Newspoll Tuesday – Small Rebounds Edition

Today’s Newspoll via The Oz comes in with the primaries running 43 (down 1) / 39 (up 2) to Labor, washing out into a two party preferred of 55/45 – a 1 point gain to the Coalition since last fortnight. This comes from a sample of 1150 1132, giving us an MoE that maxes out around the 3% mark.

On the satisfaction dynamics which hit Turbull so hard last poll, we can see a small but significant bounce back – suggesting there was a bit of overshoot in the public over the Ozcar fiasco. Net satisfaction has moved back from last poll’s 40 point drop of -33, to come in at -24. This was a result of both Turnbull’s satisfaction getting a 6 point gain to 31 and a 3 point inching back on the dissatisfaction rating to come in at 55.

Also interesting is the last 6 weeks of incremental satisfaction gain that Rudd is experiencing. Considering Turnbull’s huge personal plunge had little effect on the actual voting intention metrics, I wonder if the whole exercise did little more than rust some vote on to Labor?  Today’s Newspoll changed the Phone Poll average by a few tenths of a percent (one of the metrics we run against the electoral pendulum in the sidebar) – but made no difference to the number of seats that would change were an election held today.

Later on we’ll go over the history of Liberal Party leadership questions, incorporating today’s Newspoll question on that – but in the meantime, the usual charts come in like this.

pmsatjul13 opsatjul13

ppmjul13 netsatsjul13

marginsatjul13

3 Comments

  1. 1
    Phil
    Posted July 14, 2009 at 10:52 am | Permalink

    Does anybody do a seat by seat poll? It would be facinating to look at Curtin!

  2. 2
    Aristotle
    Posted July 14, 2009 at 4:06 pm | Permalink

    Possum, the site looks good. Plenty of stuff to chew over and nicely displayed. Good work.

    Just a query, though. You used to display LOESS regressions graphed for both parties with high and low sensitivity scenarios. I always found those graphs to be very instructive.

    Is there some reason you’ve stopped doing them? Or have I just not been able to locate them?

  3. 3
    Posted July 17, 2009 at 10:29 am | Permalink

    Phil – only in election campaigns, and then only for selected seats.

    Ari, I’ve chosen to use the pollytrend methodology rather than the old LOESS hi and low sensitivity regressions. Effectively they’re measuring the same thing, but Pollytrend just does it better.

Post a Comment

You must be logged in to post a comment.