How not to design a poll on global warming

Every month or so there is a new opinion poll saying that more people are getting more sceptical about global warming.

The latest survey from the Yale Project on Climate Change Communication continues the trend, the headline being that “Number of Americans who don’t believe in climate change rises“.  The survey report itself says “the proportion who do not believe global warming is happening has increased 7 percentage points since Spring 2013”.  This is an increase from 16% to 23%.  If we wanted to cherry-pick our start dates, we could say that the proportion has virtually doubled (from 12% to 23%) since September 2012 (see graph on page 8, shown below). There has been a corresponding decrease in the number of ‘don’t knows’,  and people seem to be more sure of their opinions, suggesting that the issue is becoming increasingly polarised.


But what exactly does “believe global warming is happening” mean? Here is the actual question asked, see page 33 of the report:

Recently, you may have noticed that global warming has been getting some attention in the news. Global warming refers to the idea that the world’s average temperature has been increasing over the past 150 years, may be increasing more in the future, and that the world’s climate may change as a result.

What do you think: Do you think that global warming is happening?

The poor survey participant is faced with three very different statements, and is asked to respond to all three of them with a single answer, ‘yes’, ‘no’ or ‘don’t know’.  I have no idea how to answer this question. I’d like to answer ‘yes’ to the first, and ‘probably not much’ to the second and third.

The press release, here at WUWT, shows that the Yale researchers are not exactly approaching the work with objectivity:  “Our findings show that the public’s understanding of global warming’s reality, causes, and risks has not improved and has, in at least one important respect, gone in the wrong direction over the past year,” said researcher Ed Maibach. On his university web page he says “His research currently focuses exclusively on how to mobilize populations to adopt behaviors and support public policies that reduce greenhouse gas emissions and help communities adapt to the unavoidable consequences of climate change.”  Perhaps most remarkable is Maibach’s belief (see press release) that it’s all just a communication problem. “Better public communication about global warming is needed now more than ever.”  Will they never learn? Clearly, Maibach and his colleagues at Yale and all the other organisations devoted to promoting climate activism to the public have just been doing something slightly wrong, and if they just increase their efforts and change their message slightly, all will be well and public opinion will start to move in the ‘right’ direction.

Maibach’s colleague Anthony Leiserowitz seems to be equally at sea. He thinks the problem is media coverage: when the media trumpeted the IPCC’s groundless claim of 95% certainty, some of them might have also mentioned the pause in warming, and this might have caused people to become more sceptical. The reality of course is that if communication is the key issue, and if people believe what they are told by the media, then the IPCC propaganda blitz since September should have decreased public scepticism about climate change.

7 thoughts on “How not to design a poll on global warming

  1. Stupid question. Clearly the average world temperature HAS increased in the last 150 years and the overall trend has been up, with various 20-25 year cooling trends. Only loonies would not believe the temperature record.

    It seems you rarely see questions like “Do you believe that global warming will lead to catastrophic consequences, within 50 years, as predicted by James Hansen, Al Gore, Michael Mann, Paul Ehrlich and others?”

    And it would be equally interesting to ask a secondary question “Do you believe that mild warming of 1 or 2 degrees fahrenheit by the end of the century would be mostly beneficial for the world’s population?”

  2. I’ve bben loking at the survey as well

    and they completley ‘forget’ to make the distinction between

    ‘global warming – natural’
    global warming – man made’

    or even ‘global warming man made and natural, and what proportion of’

    I wonder how many said they don’t believe in global warming at all –

    answered that way because they knew the surveyor wanted a yes answer for AGW but the participant, had no choices to reflect, bit of both, or natural.

    Most of the question in the survey would be impossible for me to answer, bcos of this problem

  3. The alarmist camp shoots themselves in the foot with these polls, because after asking questions that tend to tilt towards their views they then go on to believe those results are meaningful rather than the propaganda they are.

    There isn’t a small group of stubborn denialists refusing to follow the consensus, and there never has been. There has been for the last decade about 15% dreadfully worried by global warming, and 15% who think it’s complete rot. The other 70% just say whatever is required to keep the people asking them the question happy. Currently that is generally that warming is happening, because that makes them seem like the sort of people who care — and who doesn’t want to be seen to care?

    Until the alarmists start asking the correct questions in polls (or, rather, not believing the stupid ones) they won’t get a grip on what people actually believe. A sensible political party engages unaligned pollsters to test properly, because they know that their own people will inevitably tilt results.

    So we should be grateful for polls like this. They leave the alarmist flummoxed because people are steadfastly behaving “irrationally” and hence their strategies for dealing with the problem are wrong-headed. If they realised what their actual problem was they would be a great deal more dangerous.

  4. Mooloo, yes, I think you are right about the 70%. When faced with a question like
    “Are you concerned about climate change? Yes/No?”
    most people will tick the Yes box, because they know that is the politically correct, socially acceptable answer. This does not mean that they really believe it’s a major problem or that they are prepared to take any action about it.

    On asking the correct questions, see this interesting new poll from Mike Haseler. I don’t know if these are the “correct” questions or not, but some of them are quite thought-provoking.

  5. While that poll is OK for self-admitted sceptics, that’s not how you design a poll to find out what people in the middle think about global warming.

    To get to the real values, you need to write loaded questions from both sides.

    So you need to ask questions like “Do you trust industry to act morally on serious issues” because you need to sort out the anti-industry mob from the those who support industry even while worried about carbon dioxide.

    You need to ask questions like “Do you think it is OK if the Green lobby exaggerates sometimes, or should they stick to factual rebuttal?” to sort out those who trust the Greens no matter how wildly they scream.

    You need to ask what fall in income people would support in order to decarbonise the economy. That’s the crucial one for the uncommitted, after all.

    You have to ask whether a person would support a postponement of carbon reductions if the country was in a depression, or whether we should decarbonise regardless. I don’t see that one getting a lot of takers, myself.

    It would take a lot of money and time. The people with the money would not like the answers, so won’t be asking the right questions any time soon.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s