Labor was up in today’s Newspoll and down in the ACNielsen poll. So much so, these look like contradictory results.
If you are a Labor voter, you will find solace in the 53 per cent national two-party preferred vote prediction from Newspoll. If you are a Coalition voter, there is hope in the 50-50 results from ACNielsen.
Newspoll had the primary vote results as 41 per cent for the Coalition and 42 per cent for Labor.
ACNielsen had the primary vote results as 43 per cent for the Coalition and 37 per cent for Labor.
The biggest difference was the prediction of Labor’s primary vote.

Who got it wrong? There is no way of knowing for certain, but I would guess it was more likely to be Newspoll on this occasion. The recent Morgan poll and today’s ACNielsen poll are telling much the same story.
Furthermore, according to Dennis Shanahan
The seven-point rise in Labor’s primary vote is the biggest single rise for Labor against the Howard Government, overtaking three previous six-point rises after Cheryl Kernot defected from the Democrats to Labor in 1997, Simon Crean’s budget reply speech in 2004 and when Mark Latham became leader of the Opposition in December 2004.
In my view, the happenings over the last fortnight were not in the same ballpark as the earlier six-pointer moves in voting preference. None of the fundamentals changed over the past fortnight. There were no breaking stories, no personality changes and no headland speeches.
However, as I said, it was only a guess and there is no way of knowing for certain. In these circumstances, I think the best thing we can do is wait for another fortnight to see what further trends emerge.
Why did someone get it wrong? Statistics is an inexact science. Theoretically, we expect the population estimate to be outside of the margin of error one time in twenty. In practice, opinion polling shows even more variance than we would expect from statistical theory. Over time, all of the opinion-polling companies are guaranteed to produce the occasional anomalous result. As a consequence, interpreting opinion polling is as much an art as it is a science.
Did they agree on anything? ACNielsen and Newspoll had more comparable outcomes from the attitudinal polling. Both rated Howard as the preferred Prime Minister by a similar two-to-one margin. Both also had Howard’s leadership/approval rating around twice that of Beazley’s.
Aggregated monthly results. In terms of the aggregated monthly polling, March was a better month for the Coalition, with its primary and two-party predicted results improving.

Tell me more. For the Newspoll result see Dennis Shanahan. For the ACNielsen result see Michelle Grattan and Louise Dodson.
The usual opinion poll graphs are here. You may need to hit the refresh or reload button on your browser to see the latest graphs.