After the release of an Auckland Transport survey claiming two thirds of Aucklanders support more cycleways, Mike Hosking stepped into the ring to fight Statistics. Madeleine Chapman crunched the numbers.
This post was first published June 20 2018
David v Goliath. Roe v Wade. And now, Mike Hosking v Statistics. These will be the battles oft referenced in classrooms, bars, living rooms, and courtrooms for years to come. David defeated Goliath, Roe overcame Wade (for now), but who triumphed in the great 2018 head to head of Hosking and Statistics?
Well, it was close. Join us on this journey.
Monday 18 June 2018
0500 – The NZ Herald publishes a story from senior reporter Simon Wilson, detailing the results of an Auckland Transport survey on cycling. “Support for cycling overall is at 57 per cent, with 34 per cent saying they are ‘very supportive’. Those not supportive of cycling are at 11 per cent, with 8 per cent ‘very unsupportive’,” reported Wilson.
It is reported that “The survey was conducted online among 1459 respondents during April. Responses were weighted to ensure a representative spread of age, gender and location.”
These are numbers. Statistics has come out swinging.
0555 – A pair of distressed jeans enters the ring by way of the Newstalk ZB studios. Mike Hosking joins his wife Kate Hawkesby as she ends her shift and he begins his. They’ve both seen the numbers that Statistics punched into their lives and are ready to counter, like Team Rocket in Pokemon. And much like Team Rocket, they love to speak over one another.
Mike: That cycle thing… That is a rort. It is false.
Kate: Guess how many people were involved in that survey. 1500 people.
Mike: What kind of survey was it?
Kate: You can’t say two thirds of Aucklanders –
Mike: An opt-in survey, bored deadbeats with nothing to do.
Kate: Keyboard warriors ringing their aunties and uncles and cousins going ‘hey can you vote’…I refuse to believe anyone in Auckland wants more cycle lanes.
Mike: The Herald, who should know better, pass this crap off as news nowadays. And it’s wrong. It’s fraudulent.
Kate: It can’t be two thirds of Auckland.
Every good counter attack involves a bit of beneficiary roasting so Hawkesby and Hosking have done well. Hawkesby is merely a bit player in this battle but she stamps her mark with “I refuse to believe anyone in Auckland wants more cycle lanes.” Go in, sis.
0637 – Despite already knocking Statistics down by suggesting only bored deadbeats would participate in a survey, Hosking continues his assault.
“The claim is that two thirds of Aucklanders believe cycle lanes are good for the city and would welcome them into their communities. And it’s crap. It’s an opt-in survey done over a month and the response rate was, out of a population of a million and a half, it was 1459 people. And they further claim it’s weighted to ensure representation. 1459 responses between your seven regions you’re left with about 208 people per region. It’s farcical.”
0638 – Statistics are non-violent. They don’t even want to be in this fight, but they’re being pummelled by Hosking so they send in TRA, the market research company that conducted the survey.
TRA attempts to bore Hosking to death with facts such as “Respondents invited to take part in the survey were proportional to the Auckland population as per 2013 census, reflecting Age, Gender and Auckland Regions (Rodney, North, Central, West, East, South, Franklin).” and “Respondents didn’t know what the survey was on or for before completing the survey, removing any participation bias.”
The attack, though factual, is ineffective.
0639 – “They’ve got an agenda for cycle lanes and they want to pump the agenda for cycle lanes so they rig a survey, they call an opt-in survey a legitimate survey – which it’s nothing of the sort – and then they get some patsy to publish it.”
0640 – Statistics catches the ‘opt-in’ survey grenade and lobs it right back by reiterating the survey was in fact not an ‘opt-in’ survey, but a survey where people were invited to participate, ignorant of the topics that would be covered.
0641 – Patsy Simon Wilson wakes up.
0700 – Hosking swallows the opt-in survey grenade and announces his own actual, real, this-time-it-really-is-an opt-in survey.
“We’ve opened up our survey this morning on the Mike Hosking Breakfast Facebook page. Go and answer the simple question. Are cycle lanes good for the country? And we’ll see how many responses we get. They did it over a month to get 1400 responses, I suspect we’ll probably get that by about 8:30.”
If this was a fight to see who could get the most survey responses, Hosking would win. Thankfully, sampling exists for a reason.
0710 – Statistics merely shakes its head. They’ve tried to make it so easy, so that fights like this don’t happen. Yet here they are.
Statistics holds up a giant online sample size calculator revealing that a balanced sample size of 1419 people would give a confidence level of 95% and a margin of error of 2.6%.
Statistics then mentions how much Hosking loved surveys conducted with the same sampling methodology in the last election. “Another survey, two surveys like it and it’s real trouble,” Hosking said of the Very Accurate and Definitely Different From This election polls.
0745 – Hosking proudly reports that the Statistics were wrong in the AT survey because his own opt-in survey made up of ZB listeners and ZB fans on Facebook had voted heavily against cycleways.
0750 – Statistics sighs. What more is there to do but continue yelling the truth at a man with fluffy earmuffs on?
0810 – Simon Wilson eats a croissant at his desk. It’s a bit dry.
0830 – Hosking announces that his own poll has received the same number of votes as the AT survey and the result is a small 36% in favour of cycleways.
“Under their rules, our results are as legitimate as theirs.”
This is his knockout punch, showing that he alone can conquer Statistics. Statistics does nothing, just waits. The numbers always come out on top.
1400 – Hosking’s poll is now at 50/50 with just over 4000 votes cast.
1800 – With over 7000 votes cast, the support for cycleways is now equal to that shown in the AT survey, with two thirds of participants voting in favour.
Tuesday 19 June 2018
0500 – Hosking finds an unlikely ally in his losing battle: the NZ Herald. In the morning editorial, the Herald looks back on the AT survey and finds it wanting. “AT needs to be recording their use, and asking hard questions about their value, not surveying self-selecting samples online for the answers it wants.”
0501 – Statistics reiterates the survey was neither opt-in nor self-selecting.
0749 – Hosking comes out for round two with fire in his pants. He’s upset that the Herald did not report on his survey when they reported on the AT one. His producer sidekick tags in to mention the poll is now showing majority support for cycleways, but only because it was “hijacked by cycle lobby groups”. They’re both down for the count but are still kicking. Hosking goes for one last heave as the ref reaches a ten count.
“That means the opt-in survey is open to corruption thus making it not news in the first place.”
0750 – Statistics stands in the corner and calmly reminds everyone yet again that the AT survey was not an opt-in survey.
2200 – Support for cycleways in the online poll reaches 79 per cent with over 15000 votes cast. Mike Hosking still holds the strong belief that he has proven multiple points at once by owning himself with a Facebook poll. Statistics sips a cup of tea and smiles. Simon Wilson is writing about buses.
The Bulletin is The Spinoff’s acclaimed, free daily curated digest of all the most important stories from around New Zealand delivered directly to your inbox each morning.
Subscribe to Rec Room a weekly newsletter delivering The Spinoff’s latest videos, podcasts and other recommendations straight to your inbox.