Cruz wins Iowa caucuses narrowly; Clinton defeats Sanders by 0.3%

Religious right winger Ted Cruz has won the Republican Iowa Caucuses with 27.7% of the votes, to 24.3% for Donald Trump and 23.1% for the establishment favourite Marco Rubio. Ben Carson was a distant fourth with 9.3% and Rand Paul fifth with 4.5%. The other seven candidates won a combined 11.1%, with former favourite Jeb Bush winning just 2.8%. Jim Gilmore finished dead last with 12 votes out of over 180,000.

In my opinion, Rubio is the biggest winner from the Iowa caucuses. Until the final weekend, Rubio had about 15% support in the Iowa polls, but two polls taken over that last weekend gave Rubio 22% and 19% support. In the final days, it appears that Rubio was able to consolidate support from the other establishment candidates.

The turnout at the Republican caucuses was over 180,000, up from about 120,000 in 2012. A high turnout was expected to help Trump, whose voters are less likely to have previously attended caucuses. It looks as if Trump's problem was not his turnout operation, but that many also came to vote against Trump, and late deciders broke heavily against him.

While Cruz won, this was not a surprising outcome as religious right candidates Rick Santorum and Mike Huckabee also won the Iowa caucuses in 2012 and 2008. However, neither of these candidates won the Republican nomination, and Cruz's narrow victory is unlikely to give him the momentum to be a genuine chance of winning the nomination.

Trump was not expected to win Iowa until about two weeks were left, when he moved ahead of Cruz in the polls. However, he has been leading since July in New Hampshire (NH), the next state to vote on the 9 February. A NH loss would be very bad for Trump.

The Huffington Post's Pollster poll aggregate for NH currently has Trump way ahead with 34%, followed by Cruz and John Kasich on 12%, Rubio on 10%, Bush on 9% and Chris Christie on 6%. The danger for Trump is that Rubio consolidates the establishment vote following his unexpectedly strong Iowa performance.

In conclusion, the Iowa results are a blow for Trump's chances of winning the Republican nomination, but not a fatal one. If Trump wins NH next week, he will be back in the game.

Clinton defeats Sanders 49.9-49.6

In the Democratic caucuses, Hillary Clinton has defeated Bernie Sanders by just 0.3%, with only one booth outstanding. Note that the Iowa Democrats do not report raw vote totals, only the total number of delegates each candidate earned for the state convention later this year.

Sanders needed a win in Iowa more than Clinton to build momentum, so a very narrow Clinton win in Iowa is a blow for Sanders. The NH primary is next, and Sanders has pulled away to a 55-38 lead over Clinton in that state. However, the overwhelming white populations of Iowa and NH make these states strong prospects for Sanders.

After NH, Sanders will need to make inroads with black and Hispanic voters to have any chance of winning the Democratic nomination. Clinton's current dominance with minority voters make her the clear favourite to win the nomination.

Nate Silver says that polling in primaries and caucuses is very tough, and so excuses Selzer's final poll that made an eight-point error. In my opinion, it is essential to poll in the final days in such a potentially volatile situation. Two lower-rated polls that conducted the entirety of their fieldwork in the three last days before the caucuses performed much better on the Republican side than Selzer and other higher-rated pollsters that polled in the middle of the last week before the caucuses.

Little change in Australian Newspoll results

This week's Newspoll, conducted 28-31 January from a sample of 1840, has the Coalition leading by 53-47, unchanged on the last Newspoll in early December. Primary votes are 46% for the Coalition (up 1), 34% for Labor (up 1) and 11% for the Greens (down 1). Both Turnbull's satisfied and dissatisfied ratings are up a point, to 53% and 31% respectively, for a net approval of +22. Shorten's net approval rating was up three points to a still dreadful -35.

The one significant change is that Shorten now trails Turnbull 59-20 as better PM, up from a 60-14 deficit in December. The improvement for Shorten on the better PM score can probably be attributed to left wing voters who have become disillusioned with Turnbull, and thus now unenthusiastically prefer Shorten to Turnbull.

This Newspoll also asked about the GST, finding that even with a package including tax cuts and compensation for low income earners, 54% would oppose increasing the GST from 10% to 15%, and 37% would support such a change. 71% would prefer that the next election be held as scheduled late this year, while only 21% want an early election. 45% thought Abbott should retire at the next election, but 26% thought he should become a Minister, and 20% thought he should remain on the backbench.

The weekly Essential report was 51-49 to the Coalition, a 1% gain for Labor since last week. The fieldwork for voting intentions in Essential polls is conducted over two weekends, and this Essential had a sample of 1810. Other questions are based on one weekend's fieldwork.

36% supported Australia becoming a republic, with 31% opposed; in September 2015 it was 39-29 support. 45% supported a republic at the end of the Queen's reign, with 29% opposed. 56% thought a republic at some stage was likely, and 24% thought it unlikely. While there was lukewarm support for a republic, changing the Australian flag was opposed 55-33, changing the national anthem was opposed 54-28, and changing the date of Australia Day was opposed 59-23.

Over 60% had a lot or some trust in the ABC and SBS TV and radio news and current affairs. Most other media sources had between 40% and 50% trust, with Internet blogs (23%) and commercial talkback radio (34%) at the rear. All major newspapers except The Courier Mail (47%) had between a 56% and 66% trust rating; the News Ltd papers have gained trust since June 2015.

Adrian Beaumont, PhD Student, Department of Mathematics and Statistics, University of Melbourne

This article is republished from The Conversation under a Creative Commons license. Read the original article.