A Galaxy research poll on same-sex marriage came out on the weekend, commissioned by the same-sex marriage lobby, and has been uncritically reported by the Fairfax Press as finding “Rudd take on gay marriage pulls in voters”.

Of those intending to vote for the Coalition, 11 per cent said they would be more likely to vote for Labor because of Mr Rudd's support for gay marriage. Among Labor voters, 50 per cent said they would be more likely to vote Labor and 6 per cent said they would be less likely to vote Labor because of Mr Rudd's stance...

While Labor's platform supports same-sex marriage, its MPs have a free vote and many oppose gay marriage, meaning no change to the law would be possible unless Coalition MPs also had a conscience vote.

Mr Rudd last month challenged Mr Abbott to allow Coalition MPs to vote according to their conscience on the issue.

In other poll results, 58 per cent of voters, including 37 per cent of Coalition voters, agreed with the suggestion Mr Abbott was out of step with community attitudes on the issue.
But is this an accurate assessment of the poll?
A closer look at the details shows that (a) there are some serious problems with the poll’s methodology; and (b) there are conflicting conclusions which can be drawn from the poll.
Firstly, the poll undersampled Coalition voters, and since Coalition and conservative voters are much more likely to oppose redefining marriage, this skews the results towards support for same-sex marriage.
Only 344 of the 1000 voters surveyed in the poll identified as Coalition voters. This works out to less than 35%, which is significantly lower than the Coalition vote in all the other latest polls (42% in Newspoll, 44% in Galaxy, and 46% in Essential). The difference is well outside the margin of error, and shows that the poll is by no stretch of the imagination a representative sample of Australians.
In addition, the phrasing of the questions used in the poll again appears to maximise support for same-sex marriage.
There were three questions asked: 1. “Kevin Rudd has publicly announced that he believes that same sex couples should be allowed to marry. Does this make you more likely or less likely to vote Labor?” 2. “Tony Abbott is opposed to same sex couples marrying and does not allow Liberal members of parliament to vote according to their own conscience. Do you feel that Tony Abbott is out of step with community views on this issue?” 3. “Do you think that the government should change the law now to allow same-sex couples to marry, rather than wait for a national referendum?” Note that each question uses the phrase “same sex couples should be allowed to marry” or something along those lines. This is somewhat misleading, as the real topic under discussion is the legal definition of marriage, and whether it should be redefined. The issue is not whether or not same-sex couples should be able to have a wedding ceremony (they can if they want to), the issue is simply what the legal definition of marriage should be
Also, the second question is unclear: is it asking if Tony Abbott’s views on same-sex marriage or his views on a conscience vote are out of step with community views? Anyway, having a poll on whether or not a party should have a conscience vote on a particular issue is of itself disingenuous because the idea of forcing MPs vote along party lines on any issue is counter-intuitive (if you ask people if there should be a conscience vote on asylum seeker policy, for instance, most voters would likely be supportive).
We would be justified in dismissing the results of this poll based on the above flaws. But even if we leave these large problems with the poll aside, there are still conflicting conclusions to be drawn from it.
For the first question (“Kevin Rudd has publicly announced that he believes that same sex couples should be allowed to marry. Does this make you more likely or less likely to vote Labor?”) the results were divided into “much more likely” “slightly more likely” “much less likely” and “slightly less likely”.
Interestingly enough, 17% of people responded “much more likely”, while 15% responded “much less likely” (nearly the same). This shows that in terms of people whose votes are influenced by Rudd’s views on same-sex marriage, the results are almost evenly split.
In any case, the newspaper headline “Rudd take on gay marriage pulls in voters” is by no means justified by this result, as simply being more likely to vote a certain way does not necessarily pull in votes.
And on the second question (“Do you think that the government should change the law now to allow same-sex couples to marry, rather than wait for a national referendum?”) 47% responded that they wanted the government to change the law now, 30% that they wanted to wait for a national referendum, and 23% said neither.
This means that, in fact, the following conclusion could be drawn from the poll: The majority of Australians (53%) don’t want any further votes on same-sex marriage or else want a referendum on the issue. Note that this at odds with the same-sex marriage lobby’s opposition to a referendum and support for another federal parliamentary vote on the issue. That is, despite all the flaws in the poll which skewed the results towards same-sex marriage, the poll still showed that the majority of Australians disagree with the same-sex marriage lobby!
Now that’s what you call ironic.