}

Saturday, March 26, 2016

No one knows, end of story

In the aftermath of the “debate” on changing the New Zealand flag, there’ve been all sorts of declarations on social media about what the results mean. Invariably, these declarations are all based entirely on the ideology, personal beliefs, prejudices, and even wishful thinking of the people making the declarations, because when it comes to saying why people voted as they did, there’s only one truth: No one really knows.

It’s important to note first that social media has very little connection to real-life voting behaviour—if it did, then Red Peak would be the new flag of New Zealand. Instead, it came in third in First Preferences in the first flag referendum, and was eliminated entirely in the third count. Social media also told us that “masses” of people would spoil their ballots in protest, but it ended up being less than10% of the total in either the preliminary or final results of the first referendum, (and insignificant in the second). The point is, one must take unsupported opinions on social media with considerably more than a single grain of salt (though saying “a salt mine” is overstating it a bit).

The most common comment I’ve seen has been about how the Māori electorates voted overwhelmingly against change. Some have claimed that Māori voters love the current flag, others that Māori voters felt left out of the process, or resented the alternative because it didn’t have Māori symbology. It’s probable that any/all of those things may have been at play for at least some Māori voters.

However, it's equally possible that Māori voters, like plenty—perhaps most—of all voters cast a protest vote, maybe against the selection system, maybe against the cost, maybe against John Key personally, maybe ALL of these things. The problem we have is that because there's been virtually no attitudinal polling, as usual, we really have no idea why anyone voted as they did. That's goes doubly for Māori voters who are generally not polled specifically at any time.

UMR Research polled twice on the issue of the flag change, and found little difference in their second poll, indicating that most voter intention was consistent. While they didn’t get the final vote percentages correct, their margin was close—and much closer than others I saw in February. Back then, most polls had the alternative design losing by nearly 3 to 1, while UMR’s was roughly 2 to 1. In the end, of course, it was much closer.

UMR also suggested—correctly, I think—that those most likely to change their vote would be among the 20% of respondents who said they planned on voting for the current flag, but who agreed with the statement, "in principle I would like to change the flag but I just do not like the alternative.” It looks like some of those may have indeed switched, and the undecided may have mostly gone for change, thereby tightening the margin.

UMR released more detailed vote analysis than any other pollster, which gives us some idea of who was voting which way, and it gave us some insight into why. For example, they reported that two-thirds of respondents agreed with the statement, “The flag referendum has been a distraction and a waste of money. New Zealanders should send John Key a message by voting for the current flag.” 79% of younger voters agreed with that statement, UMR said, as well as that they are “more likely to be Labour and Green voters than the population as a whole” (though they didn’t release specific data to back up that statement).

Adding all that up, we know quite a lot about the stated intentions of voters aligned with various party affiliations, voters in various age bands, and male and female voters—but we know nothing whatsoever about voting intentions of people in different races/ethnicities or economic classes. This data may have been recorded but not included in what was already very complicated reporting, but maybe it wasn’t even sought.

In the absence of such publicly available data, we have no idea why voters in Māori Electorates voted so overwhelmingly for the current flag. We can draw inferences based on what was reported—such as what I’ve highlighted, and also the fact that Labour Party supporters backed the current flag 73% to 19%, and Māori voters have in the past strongly supported Labour—but those are only inferences, albeit ones at least based on publicly available data.

It seems to me that some opponents of change may have been trying to make a moral justification for what was a largely political decision made by many opponents to “send John Key a message”. That desire for justification is completely understandable, and it happens all the time in all elections and (maybe especially) referenda. But assertions made without evidence are nothing more than opinions. I can’t prove that my assertion about their motivation any more than they can prove their assertions about why Māori voters voted as they did.

The “debate” on changing New Zealand’s flag was truly awful (a topic in itself), and in the post-vote analysis we shouldn’t allow unsupported personal viewpoints to continue to squeeze facts out. I offer no opinion on why the Māori Electorates (or any other ones) voted as they did because facts matter, and I simply don’t have any beyond what polls were showing. I think we all need to be careful about reading the results as validation of a particular perspective we may hold when we have absolutely no way to prove we're correct.

Why did voters vote the way they did? No one knows, end of story.

No comments: