With the election safely out of the way, we might have expected the heat to have gone out of the great blogosphere-versus-The Australian opinion poll wars. Turns out Dennis Shanahan has other ideas:
This week’s most eye-catching figures were Kevin Rudd’s 70per cent and Brendan Nelson’s 9per cent on the question of who would make the better prime minister, a Newspoll record high for a prime minister and a new low for an Opposition leader. After The Australian put the story, which I wrote, on the front page, it captured public attention and was reported, commented upon and retold in newspapers, radio, television and blogs. As Possum Comitatus said …: “While records are meant to be broken, this one was obviously meant to be smashed. Brendan Nelson has stormed into the worst preferred prime minister result in the history of Newspoll with an astonishing 9 per cent.” Peter Brent’s Mumble and William Bowe’s Poll Bludger, sites that panned the Newspoll reporting in the past, covered it without personal comment … Yet there was one key point missing from all the commentary that has previously cropped up in analysis of Newspolls: in Possum’s words, Nelson “stormed” to his rating by 2 percentage points. Rudd’s record on preferred PM was also reached by a rise of 2 percentage points. The margin of error for the Newspoll survey on a sample of 1140 is 3 percentage points. The leaders “stormed” to these records with movements of less than the margin of error. In the past, The Australian has been castigated for reporting movements of 2 per cent and placing stories on page one based on “record” lows … Statistical bloggers forever complain about reports of movements of less than 3 per cent and essentially want polls to be banished from newspapers and public debate except during an election. On this occasion, as on previous occasions, the simple news judgment was made in writing the story and placing the story, that a record, however it is attained, is newsworthy. The bloggers thought so, as they trawled the records to find Crean’s lowest reading in Newspoll and talked about the importance of the preferred prime minister figure for leaders. If Nelson’s preferred prime minister rating drops one point to a record low of 8 per cent, is that worthy of page one again? Or do we ignore that as being within the margin of error? Fat chance. Polls interest people, influence politicians and should be treated consistently.
For the record, I personally had very little to say last year about Shanahan’s Newspoll reporting. This was partly because the subject was being done to death elsewhere (not least in my own comments threads), but also because I had more sympathy than some for the idea that Labor’s bloated lead would indeed feel the effects of gravity before polling day. My post on Tuesday’s Newspoll even managed a sarcastic dig at those who paint him as a Coalition stooge.
In other news, the AEC has commenced a redistribution for Tasmania, it having gone the maximum seven years without one. The AEC’s figures respectively put enrolment in Bass, Denison and Lyons at 1.2 per cent, 1.6 per cent and 2.3 per cent below average, with Braddon and Franklin 1.5 per cent and 3.7 per cent above. So the redistribution will presumably involve a transfer of territory from Franklin to Lyons, which is unlikely to make much difference to anyone’s electoral prospects. Changes to the more sensitive Bass and Braddon are likely to be negligible. Uniquely, Tasmanian boundary changes have effect at both federal and state level.
UPDATE: Shanahan’s central contention, that Nelson’s drop from 11 per cent to 9 per cent was within the margin of error, is questioned by David Walsh and Unicorn in comments. The latter tells us that the sampling error depends on the uniformity of the population, so the 3 per cent figure assumes a 50-50 response like you roughly get from a two-party split. Whereas the question of Nelson’s approval or non-approval in fact splits about 10-90, producing a margin of 1.7 per cent.