New Zealand's deployment of troops to train the Iraq Army continues to have strong public support, says the latest Herald-DigiPoll survey.
In the poll, 59 per cent of respondents said they supported the deployment of 121 troops to train Iraqi soldiers in the fight against Islamic State - up from 57 per cent in May. Just over one-third said they did not agree with it - the same level as in May soon after the troops arrived in Taji.
Prime Minister John Key said the level of public support reflected the view that New Zealand had to play its part in the world and help the people of Iraq defend their own country. He did not believe high public support would justify deepening New Zealand's involvement.
"Like every response, there's a proportionate response and I think we're in the right place."
AdvertisementAdvertise with NZME.
Countries, including Britain and Australia, are now revisiting the possibility of greater military intervention in Syria as well as Iraq to try to drive back Islamic State and stem the exodus of refugees.
Mr Key said it was a difficult issue because while military involvement could mean progress in the short term it risked building long-term resentment. He said that was why New Zealand had opted to train Iraqi soldiers, rather than get involved in direct combat.
Labour leader Andrew Little said Labour stuck by its decision to oppose sending the training troops, saying New Zealand's efforts were better spent on humanitarian efforts.
In defending the decision to send the troops to Iraq Mr Key gave an impassioned speech in Parliament saying he was not willing to stand by while atrocities were being committed on the people of Iraq and Syria by Islamic State.
Those words were thrown back at him by Opposition parties over the past week after his initial reaction to the Syrian refugee crisis, which was to wait until a review next year before deciding whether New Zealand would do more.
This week Mr Key bowed to that pressure and announced New Zealand would take in an extra 750 Syrian refugees.
The poll of 750 eligible voters was taken from 14-24 August and has a margin of error of +/- 3.6 per cent.