SMH and Age end relationship with polling company Ipsos after election blindside
Nine’s newspapers have ended their relationship with polling company Ipsos after the election delivered a radically different result to that anticipated by the pre-election polling.
Mumbrella understands Nine News’ separate agreement with Galaxy Research is not affected by the decision made by The Age’s and Sydney Morning Herald’s management.
Ipsos was consistently polling a Labor win prior to last weekend’s result, with the final poll published on Friday showing a national two party vote of 51-49 for Labor.
The announcement the relationship was coming to an end was made by Sydney Morning Herald and the Age national editor Tory Maguire in a column published today.
Maguire did flag that Ipsos had been predicting a low primary vote for Labor leader Bill Shorten, but states in her column those warnings didn’t excuse the fact the pollster delivered an incorrect overall forecast.
“Polling companies are the main reason Saturday night’s result took voters, the media and many political operatives by surprise,” says Maguire.
“The implications of our major pollsters making the same mistakes in a consistent way are serious.”
The column goes on to say that both the Sydney Morning Herald and the Age have to make decisions within their newsrooms about whether they will return to polling in the future. Maguire states the editorial teams were already reconsidering the use of polling prior to the election result.
Ipsos has released a statement saying it would be reassessing its own polling methods over the coming weeks.
“The Federal election was very close in terms of overall party support but Ipsos acknowledges that its final Ipsos Poll numbers clearly did not reflect the eventual Federal election results last weekend and it needs to do better,” said the statement.
“Ipsos will be conducting an assessment of its polling over coming weeks with the local team and the global Ipsos polling experts in order to understand what the research company could have done to achieve a more accurate outcome. No doubt other polling companies who covered the election will be doing the same.”
News Corp has responded to a request for comment from Mumbrella advising they will not stop reporting on polling or working with Newspoll.
Can someone explain to me why statistics is broken?
This wasn’t one poll. This was continued polling all delivering a consistent result.
So what part of the algorithm for statistical extrapolation needs updating?
User ID not verified.
I’m sure more will come out but there’s some combination of:
1) Fewer and fewer people have landlines and those that do are a small range of demographics
2) Swings that aren’t uniform
3) Small parties causing havoc with the results
User ID not verified.
For the life of me I cannot remember the name of the company, however they monitor social media posts and correctly called the election.
User ID not verified.
I’d imagine the weaker primary vote of major parties is a factor since it amplifies the volatility from place to place. Two party preferred aggregates breakdown the more the primary vote is split.
No doubt however that landlines are a factor as is the tendency with digital phones to reject unknown callers.
Parties focus on boots on the ground and social media, which says it all really.
User ID not verified.
Sounds like a convenient excuse/cover for cost cutting…..
User ID not verified.