Today brings about a Nielsen that has the tragics spinning with the Coalition hitting the lead on a two party preferred split of 52/48. The poll came from a sample of 1356, giving us an MoE that maxes out around the 2.7% mark. The primary votes broken down into gender and geography as well as the preference flows, including their change from the previous Nielsen poll come in like this:
As we can see, Nielsen – like the previous Newspoll – suggests the gender gap has changed, but unlike Newspoll, it also suggests that substantially more women are now voting for the Coalition than for the ALP – the first poll of the campaign to suggest such a thing. At this stage, it’s worth looking at the results of all the polls we’ve had since the 1st July (click to expand):
Over the last week, there has definitely been a movement back to the Coalition, but Nielsen certainly appears to have overshot relative to other movement we witnessed from other pollsters over the same approximate period. Using our Pollytrend to cut through the uncertainty, it comes in like this:
Our phone pollster trend, which includes today’s Nielsen result, is currently sitting on an ALP two party preferred of 51.3%. Galaxy, Morgan’s phone poll and Nielsen were all in the field last week with overlapping periods. Morgan got a 53% TPP to the ALP over the 27th and 28th, Galaxy got a 50% TPP on the 28th (and maybe on the 27th as well – I can’t find if Galaxy was in the field for one night or two) and Nielsen got a 48% TPP over the 27th, 28th and 29th. So Nielsen is the odd one out here, suggesting a bit of overshoot – and probably combined with the ALP week getting worse as it went on.
However, even if Nielsen went a few points too far, it did bring in results consistent with the events of last week. If we track the Nielsen two party preferred by gender and geography, a few things stand out:
Firstly, non-capital cities dropped 8 points on the TPP for Labor in a week. That doesn’t actually happen in the real world, so it’s a bit full on – but just because the magnitude of the change might be out by a few points doesn’t mean the direction of the change is wrong. Similarly, the female vote collapsed for Labor on the two party preferred, dropping by a massive 9 points in a week. Again, that doesn’t happen in the real world, but it doesn’t mean that females didn’t move back towards the Coalition in solid numbers this week – just probably not this solid.
The mediascape over the last week was filled with generically negative coverage of Labor – most of it self-inflicted – but of particular focus was the allegations that Gillard questioned the pension rise and paid paternity leave.
The latter would certainly help explain some female movement, while the pension issue would have been expected to play out more in regional Australia than the capital cities since regional Australia is older than its metro counterpart and there’s a higher level of pension reliance among that older cohort in regional Australia.
If we track the Nielsen two party preferred by age cohorts (and I’ve combined the 18-24s and 25-39’s to give us a tracking sample for each cohort with an MoE somewhere around the 5% mark) it adds some more evidence:
Young people came off a bit of a peak and 40-54s didn’t really move, but the over 55’s collapsed for Labor, from 46% last week down to 38% this week. So there would appear that there was movement towards the Coalition, and that the movement occurred in the places that last week’s events would easily explain – but the magnitude of the movement is probably slightly out considering the poll results that came from the field at the same time.
The problem for Labor of course is that the trends might continue – turning this Nielsen from what is more likely a slight outlier on the day, to an early adopter in hindsight – particularly if Labor can’t consolidate its female vote and pull back a few oldies as well.
Later today or tomorrow, we’ll have a look at the rest of the Nielsen poll, the full demographic tables of which can be seen here.
More interesting though is a 3 poll average of that table, which can be seen here – this latter table actually looks pretty spot on in terms of what appears to be the current state of play.