Via the Fairfax press, today’s Nielsen comes in with the primaries running 42 (steady)/ 37 (down 2) to the Coalition, washing out for a two party preferred of 50/50 – a 1 point gain to the Coalition since the last Nielsen in April. The Greens are on 13 (up 1), while the broad “Others” are on 8 (up 1). The Nielsen poll comes from a sample of 1400, giving us an MoE that maxes out around the 2.6% mark.

This is probably what the last Newspoll should have looked like – with a much more realistic “Others” vote and an ALP primary not floating around in Crean territory.

The recent behaviour of the approval rating results for Nielsen pretty much mirrors what we’ve seen on the Newspoll satisfaction ratings – regardless of which way you ask the question, voters aren’t happy with the PM, as we can see by running them together:

appsvsats

But wait – there’s more. If we track the primary and two party preferred vote of Labor over this year, we see the same recent plunge (click to expand).

alpprims2010 alptpp2010

We don’t usually see these types of dramatic changes in the polling unless we either have a new leader (but Abbott has been here for months), a scandal (the only scandal was over at Fox News with Words, but everything is a scandal to The Oz, what would have made the public pay attention this month?) or someone gets hit in their living standards.

What happened between mid April and early May on the latter?

Smokes went up.

Between 15-20% of the population smoke, so the sharp drop could be explained by around 1 in 4 smokers changing their vote over the tax hike. Alternatively, it could be a point or two decline from the ALP over the broader generic issues of the day and 1 in 5 or 1 in 6 smokers changing their vote on the tax hike.

Either way – some significant chunk of people moved together against the ALP in a substantial way that has been picked up across all pollsters, with phone pollsters picking up a larger movement.But they did so in such a way that Abbott’s metrics on approval and preferred PM  haven’t really moved (see the charts below), suggesting it was something the Rudd government did rather than anything Tony Abbott has done.

Smokers tick the boxes for the additional sharpness of that drop in both the ALP primary vote and the larger drop in Rudd’s personal metrics.

We’ll go through the additional Nielsen questions in a separate post later on today We’ve gone through the additional Nielsen questions on the ETS and Trust over here:

Meanwhile, the usual charts come in like this:

pmapprovmay opapmay

pmnetappsmay ppmmay

UPDATE:

Demographic tables of the Nielsen poll are available here.

Greensborough Growler in comments asked about interest rate changes and the two party preferred vote. This is what the Coalition vote vs the cash rate looks like over the Rudd government’s term.

TPPvCashrate

Something that’s popped up a few times today is whether this Nielsen result justifies the last Newspoll result or whether it was still an outlier. The problem with the last Newspoll came in the primaries – an extremely low ALP primary (35) and an extraordinarily high “Others” vote (12).

The ultimate two party preferred result of the last Newspoll (51 LNP) was arrived at by accident rather than design – as a result of the preference flows of “Others” at the last election being applied to a voting estimate of those “Others” that was 70% higher than the broad “Others” achieved in 2007.

So while the TPP was in the ballpark, the numbers that created it’s foundation  – especially the Others primary vote- have not been replicated by any pollster.

So yes, it still looks like it was an outlier (or rogue), but the two party preferred it created was still accidentally in the ball park of what we are seeing with the other phone polls such as this Nielsen and Morgan’s small sample phone poll from last week.