There is also an art of making opinion polls say what you want them to say, e.g. by asking a biased question, sampling a demographic known to skew a certain way or using a method that largely may exclude certain demographics. (E.g. internet polling)

On top of that, there will be selection bias in what polls are published. If the poll you've paid for does not support your narrative, there is no reason to publish it.

Polls are therefore often skewed one way or another, and should probably only rarely be used as support for what the general population wants. At least when the margins are small. Especially if paid for by an organization with stakes in the question.

Don't forget:

- voter suppression[1]

It doesn't really matter what percentage of the public favors X, when people who might vote for X are systematically blocked, coerced, or dissuaded from voting. And when one party tends to benefit from low voter turn-out, as is the case in the USA, it should be no surprise when very popular politicians or proposals that poll well, fail due to voter suppression.

1: https://www.votingrightsalliance.org/forms-of-voter-suppress...

One party doesn’t consistently benefit from low turnout, not anymore. Trump won almost exactly half of new voters in 2020. Also, “voter suppression” is a myth, just like “voter fraud.” Isolated instances but studies consistently fail to show any material effect. (E.g. no effect from voter ID: https://www.vox.com/policy-and-politics/2019/2/21/18230009/v...)

> > Are opinion polls actually representative of what public opinion is

>Yes.

No.

According to "Yale Climate Opinion Maps 2018", 69% of people polled in washington state supported taxing fossil fuel companies. In the same year, a ballot initiative to enact carbon taxes failed 43% to 57%. Granted, they're not asking the same question, but it seems doubtful that 26% of people were for "taxing fossil fuel companies" but were against "pollution fees on sources of greenhouse gas pollutants ".

[1] https://climatecommunication.yale.edu/visualizations-data/yc...

[2] https://en.wikipedia.org/wiki/2018_Washington_Initiative_163...

That measure specifically exempted fossil fuels.

According to that wiki link [1]

> Others opposed the measure because section 9(c) specifically exempted "Fossil fuels directly or eventually supplied to a light and power business for purposes of generating electricity" from the carbon tax.[3] This meant that coal, gas, and diesel power plants would not directly be responsible for paying the carbon tax.

Also, midterms have a much lower voter turnout and in Washington it was 58.9% in 2018 vs 75.71% in 2020 [2]

[1] https://en.wikipedia.org/wiki/2018_Washington_Initiative_163...

[2] https://ballotpedia.org/Voter_turnout_in_United_States_elect...

> That measure specifically exempted fossil fuels.

This statement is highly misleading. Fossil fuels are exempted, but only for a limited number of circumstances. The wikipedia article you linked only mentions fossil fuels being used for electricity generation being exempted. Furthermore, you failed to quote the rest of the paragraph, which mentions fossil fuels make up a minor part of electricity generation in Washington.

"However the majority of electricity generation in Washington state is derived from renewable sources. [...] The same analysis further indicated that only 4% of Washington's energy comes from burning coal, [...] Natural gas currently accounts for only 10% of Washington's energy generation, according to the Washington Post analysis. "

>Also, midterms have a much lower voter turnout and in Washington it was 58.9% in 2018 vs 75.71% in 2020 [2]

Except that the 2018 ballot measure wasn't the first time Washington tried instituting a carbon tax. They tried 2 years before (not a mid-term year) and that failed even harder, 41% to 59%.

> The wikipedia article you linked...

It is YOUR wiki link, but with '#Opposition' at the end to move directly to that section of the article...

Edit to add:

> This statement is highly misleading.

Kinda like the post before that, you said:

>>> 69% of people polled in washington state supported taxing fossil fuel companies.

and

>>> but it seems doubtful that 26% of people were for "taxing fossil fuel companies" but were against "pollution fees on sources of greenhouse gas pollutants ".

The whole reason I read through that wiki You linked, was because this did not seem like the same thing, and a lot of people with only a quick reading (such as myself) who this issue is not their main focus is only going to see as loosely related, if that - If Im the 69% that wants to tax fossil fuels, and the ballet says taxes on "greenhouse gas" - Im not going to care or be very happy with the measure, because I agreed I want to tax "fossil fuels"

As for:

>> Except that the 2018 ballot measure wasn't the first time Washington tried instituting a carbon tax. They tried 2 years before (not a mid-term year) and that failed even harder, 41% to 59%.

I was responding to your post and that wasn't included - and it could just mean that as the years progress, people are getting more concerned..who knows

>If Im the 69% that wants to tax fossil fuels, and the ballet says taxes on "greenhouse gas" - Im not going to care or be very happy with the measure, because I agreed I want to tax "fossil fuels"

Again, I'm skeptical whether this distinction is responsible for a 26 percentage point drop in support. Fossil fuels make up the overwhelming majority of greenhouse gas emissions in washington state. Taxing greenhouse gasses is largely synonymous with taxing greenhouse gasses. Outside of a very small group whose livelihood involves non-fossil fuel greenhouse gas emissions (cow farmers?), I'm having a hard time imagining why you'd be against a greenhouse gas tax but would support a fossil fuel tax.

> I'm having a hard time imagining why you'd be against a greenhouse gas tax but would support a fossil fuel tax.

Where did I say that?

I was pointing out that choice of words matter. This isnt even about the tax, the original point was about if what people say in polls is real - you argued no and presented very shaky evidence - I was pointing out the flaw in it so you could refine your position or rethink it - what you choose is up to you - but thinking all 69% of the people who would want to tax “fossil fuels” and are also able to make that connection to “greenhouse gases” or not just really dislike fossils fuels for other reasons, I think may be too generous. Then add the people who dont think it goes far enough, plus the midterms and you have a not great argument.

Either way, I havent stated any actual opinion on the tax matter nor have I thought about in the discussion in any way other than to point out what was missing from your example about polling opinion - the words matter, and also voter turn out

The first two do not seem different for polling vs. voting in a referenda.

That's not really true because referenda usually become partisan. One of the parties is often the one initiating it and choosing the verbiage, which means it's often presented extremely misleadingly, too.