The Sunday Mail today carries a big sample ReachTEL automated phone poll of state voting intention in Queensland, tailored to produce samples of a bit over 1000 for each of South East Queensland, North Queensland and the remainder of the state (3636 respondents overall), notwithstanding that the first of these accounts for by 63% of the state’s 2015 election voting population, assuming we’re using the same definitions. The result is a good deal worse for Labor than the last published state poll, which was conducted by Galaxy in late April, with primary votes after exclusion of the 6.7% undecided at 31.9% for Labor, 35.3% for the Liberal National Party, 17.0% for One Nation and 9.4% for the Greens.
The report cites primary vote swings against Labor of 13% in north Queensland, 9% in the regional remainder and 5.5% in the south-east, compared with respective results of 4.3%, 7% and 6.6% for the LNP – although I’m having a particularly hard time accounting for a definition of north Queensland consistent with such a big swing against Labor compared with the 2015 result. The headline two-party figure, based on respondent-allocated preferences, is 51-49 in favour of the LNP, compared with 52-48 to Labor in the Galaxy poll. A graphic in the report identifies a 51.5-48.5 in favour of Labor on the question “which of the following two parties would you preference higher” – this would be the respondent-allocated preference prompt for those who chose a minor party. Given the shaky record of respondent allocated preferences, it would be very difficult to predict how an election would play out in practice with non-major party support at the levels indicated.