Making sense of the polls

Making sense of the polls

Big move to Coalition in Ipsos not replicated in other polls

This week’s Ipsos gave the Coalition a crushing 56-44 lead using previous election preferences, a 3% gain for the Coalition since the last Ipsos in mid-October. However, the other two polls released this week had no movement on two party voting intentions, while primary votes suggested slight movements to Labor that were lost in the rounding. Here is this week’s poll table.

polls mid Nov.

Morgan had about a 1% pro-Labor skew under Abbott, but has developed about a 2% pro-Coalition skew under Turnbull. The last four Morgan results have all been 55-45 to the Coalition by previous election preferences while other polls had the Coalition on 52 or 53% Two Party Preferred (2PP). So this Morgan result should be read as 53% 2PP by another pollster.

Both Morgan and Ipsos asked for respondent allocated preferences, and both polls had the Coalition doing 1% better on this measure than on previous election preferences. Under Abbott, Labor usually did better on respondent preferences. The change probably reflects hard right voters who have defected to Others going back to the Coalition on preferences, and Greens voters being slightly more Coalition-friendly under Turnbull.

In Ipsos, 69% approved of Turnbull (up 1), and only 16% disapproved (down 1), for a net approval of +53. Shorten’s net approval was -28, down four points, and his lowest rating in Ipsos/Nielsen history. By 66-28, voters opposed increasing the GST rate; this compares with 59-37 opposed in April. However, if a GST increase were accompanied by tax cuts and compensation for those earning under $100,000, a GST increase is supported by 52-41. Note that if the voting intentions in this poll are too strong for the Coalition, the leaders’ ratings and GST questions will also be affected.

The terrorism in France occurred on Saturday morning Melbourne time, but appears to have had little impact so far. Only one third of Ipsos’ fieldwork was on Saturday, so the 56-44 result cannot be attributed to terrorism. Morgan’s fieldwork included both Saturday and Sunday last weekend, but recorded no change.

While terrorism may not have impacted this week’s polls, it would be unsurprising if the Coalition’s lead increased in coming weeks. If the polls move towards the Coalition, it will be difficult to say whether this Ipsos was indeed an outlier.

The Ipsos poll has given the Coalition a sizable jump in the poll aggregates. Kevin Bonham’s aggregate is now at 53.6% 2PP to the Coalition, a 0.8% gain for the Coalition since last week. The Poll Bludger’s BludgerTrack is at 54.4% 2PP to the Coalition, a 0.9% gain for the Coalition since last week. Primary votes are 46.3% for the Coalition, 30.7% for Labor and 11.2% for the Greens; Labor has lost 1.6% on primary vote since last week, with Others gaining 0.8%.

More on Essential

In this week’s Essential, Labor’s attributes showed little change since September 2014. The Liberal Party had a 16-point increase on “divided”, a 10-point increase on “has a good team of leaders” and a 7-point increase on “moderate”. When the two parties are compared, Labor is perceived as far better on the interests of working people, and the Coalition far worse on being too close to big business. The Coalition leads Labor by over 10 points on “good team of leaders” and “vision for the future”.

29% thought the Coalition government is too soft on asylum seekers, with 25% for too tough and 31% for the right approach. In April, these figures were 27% too soft, 22% too tough and 34% about right. By 54-31, voters agreed with off-shore detention of asylum seekers on Nauru and Manus Island.