A blog about advertising, copywriting, creativity &c.
How Polling Language Can Mess With Your Head

How Polling Language Can Mess With Your Head

(Content note: This isn’t a post about politics, but it definitely has politics, and you can probably guess what kind. The “close tab” button is the little X at the top.)

A photo of a kitten peering over the edge of a battered black button with a red handle, against a dark, blurred background.
I don’t know why this photo came up in a search for “liar,” but I can only assume this kitten has really pissed someone off.

I’ve already blathered about the use of statistics to throw you off and change the way you think, and why understanding them helps you understand the world and protect yourself from statistical BS you may encounter. (I’m sure you found the post scintillating.) Naturally, it’s time for words to enter the chat (that being, of course, my stock in trade).

The ability of words to influence your choices and change the way you think is basically how we pay our rent. And that might give us a little hubris, thinking we’re immune to the way other people can use words against us. And not just in words directed at us, either.

Okay, this is getting a little abstract.

Polling and Questions

In polling, the way questions are asked can have a major impact on the results of the poll. Obviously. But a recent batch of polls has been a great illustration of that. Following the ICE killing of Renee Good in Minneapolis, various outlets have been seeking public opinion on the topic, with various results (and here I credit the NYT with pointing me in this direction in the first place).

Napolitan News Service/RMG Research

There’s a saying about bovine excrement in proximity to the word “but,” but I think in terms of the All-Cleansing “But” — making a caveat before the word “but” that absolves you of anything you say after. I’m not a dog-hater, but I do think all corgis should be set on fire. The question in question uses “however” rather than “but,” but the effect is the same:

Everyone agrees that the death of a woman in Minneapolis was a terrible thing. However, some say that she was mostly responsible because she had been acting to obstruct ICE operations and was threatening toward ICE agents. Others say that the ICE agent acted irresponsibly and was never really in any danger. Which is closer to your view?

A photo of a beige alpaca lying in the grass, not flat-out but tummy-side down, with his head up, looking extremely content.
This alpaca is also clearly lying and y’all, this month has been a long year and I’m going with it.

There’s our implied “but.” The sentence before “however” eases us into the subject gently, giving everyone a break on having to have an opinion on whether Renee Good’s death was, in fact, terrible — obviously everyone thinks it was. And then “however” clears everything out so we can move on and look at it the question of blame from a more analytical, pragmatic standpoint. Human element before “however,” pragmatic assignment of culpability after.

Is one way of asking a poll question about it.

CBS News/YouGov

The other version of the provided question skips the appetizer and goes straight for the entree.

From what you know, do you think the shooting of the Minneapolis woman by an ICE agent was justified, or not justified, or is it too soon to say?

Boom. Yes or no. (Or [wordless shrug].)

Yes, there’s no “but,” All-Cleansing or otherwise, but there’s another significant difference between the two questions (did you catch it already?): They’re not actually asking the same thing. The first asks about who’s responsible, and the second asks about what’s justified. The first looks at the individuals involved and their actions, and tries to assign culpability. The second looks at the overall circumstances and asks whether what happened should have happened. Taken together, of course, they’d create a false dichotomy, which we’ll discuss below and is why you’ll rarely if ever see them taken together.

The NYT also points out that polls like these can be a situation where people don’t take time to parse the question anyway and just give their opinion on the overall ongoing situation — a vote for the red team or the blue team.

In saying all this, I’m not trying to imply these polling services are (or aren’t…) intentionally trying to sway results by angling their questions. Crafting a completely neutral question is a job of work. And the services can’t control how the data is used and interpreted once it’s out of their hands. News/opinion outlets with any kind of editorial slant are likely to choose the poll results that reinforce their existing leanings, and people with political leanings are likely to do the same.

(And then there’s Caperton, who despite her obvious political leanings is there crying, “But it’s a false dichotomy!”)

(And, hey, let’s talk about those.)

False Dichotomies

A photo of a woman lying face down on the sidewalk, flat out with her arms above her head. She’s wearing a white tank top and short blue shorts, her blonde hair is in a ponytail, and she’s absolutely covered in birdseed and surrounded by dozens of pigeons eating said birdseed.
Girl, same.

Webster’s Dictionary defines — totally kidding. I would never. A false dichotomy is a logical fallacy that presents just two possibilities when, in fact, tons of possibilities are available. Like the one I just mentioned above — a person could believe, for instance, that Renee Good was mostly responsible for being threatening (as a woman in a slow-moving Honda Pilot naturally would be), but it still wouldn’t make the shooting justified.

It would be stating the absurdly obvious to say that in polling and in life, you’ll be presented with either-or choices that do not compute. (Ask any autistic person about “Would you rather go to the theater or a museum?”) Now, at least, you know the name for it, if you didn’t before. And when you recognize it in a binary poll, you can know to take the results with a grain of salt. And when you recognize it in conversation, you can say, “That’s a false dichotomy!” and people will think you’re smart.

To that end, Data for Progress took a look another question: of whether to “abolish ICE.” Because… how? In what manner? How much? To what end? In DFP’s study, they didn’t even ask the binary question but look at all the factors that could influence it — including the question,

When some elected officials say “Abolish Ice,” referring to U.S. Immigration and Customs Enforcement, what do you think they mean?

Clarity is crucial, and false dichotomies are a trap, and “What do you mean by that?” gets a seriously bad rap in my book. Also, we love it when you include your survey mythology at the end of your study, <3.

Watch Your Words

(For the record, the answer is to the above question is, “Is it a movie theater or a theater-theater? What’s showing? What time of day is it? Will there be kids there? What kind of museum? Will there be places to sit down? Is there a special exhibit going on? Will there be kids there? Is there one of those guided tours with headphones?” but I’ve never seen that on an autism assessment.) 

So, yeah, now you can get super neurotic about every question ever asked in your presence. And now, instead of calling bullshit on your friends, you can say, “False dichotomy!” and they’ll definitely think you’re cool. But in all seriousness, I do hope you hold onto the All-Cleansing “But.” Not all “But”s are All-Cleansing, and when you start to recognize them, you start to learn new things about the world around you.

Go forth wiser and more aware of more ways words can be used, intentionally or unintentionally, consciously or subconsciously, covertly or obviously, to mess with your dome.

Leave a Reply

Your email address will not be published. Required fields are marked *