In this second exchange (1st here) between Adam Corner (Talking Climate blog) and Geoff Chambers – (a regular and prominent commenter at several climate sceptic blogs), they continue to discuss research on the psychology of scepticism.
Comments are very welcome – but please be aware they will be tightly moderated for civility to ensure all comments are on-topic (some leeway on topic if interesting/relevant).
Is this kind of initiative useful? Should it happen more often? We look forward to hearing your thoughts. (Talking Climate)
In your paper (Corner, Whitmarsh & Xenias, 2012) which set off our discussion, I noted that you administered a battery of questions called the New Ecological Paradigm (NEP) to your respondents. It didn’t affect your research, but it got me interested in the whole question of how you assess environmental attitudes.
The NEP has its origins in a 2000 paper by Dunlap et al. It updates a previous set of questions called the New Environmental Paradigm developed by Dunlap and van Liere in 1978. The paper, with a list of the questions and analysis of a pilot test can be found at [here] The NEP seems to me to raise the kind of questions about environmentalism, and in particular research into attitudes to the environment, which Ben has been examining at Climate Resistance. What’s your take on it, and why did you include it in your research?
We included it in our research in order to have a well validated scale (i.e., used lots of times and found to have good internal consistency between the items) for measuring people’s beliefs about the environment. We wanted to know how they related to people’s beliefs about climate change (which we measured using Lorraine Whitmarsh’s scale for ‘climate scepticism’), and in turn, how NEP scores might determine people’s ratings of the newspaper editorials.
As you might expect, people’s score on the NEP scale was a significant predictor of their scores on the scepticism scale – the more ‘pro-environmental’ they were (as measured by the NEP), the less sceptical about climate change they were, over and above who they voted for, and whether they were a member of an environmental organisation.
We did not do a great deal with this analysis, but I think it helps to set the scene for what is going into a sceptical judgment about climate change: in part, it is coming from disagreeing with the NEP scale. So, in part, scepticism about climate change is coming from disagreeing with items that measure ‘pro-environmental’ beliefs (and measured them long before climate change was known about by the public.
The NEP is described as measuring whether people have an ‘ecological worldview’ and a ‘pro-environmental orientation’. I don’t see any major issues with this as a description, but you could throw various criticisms at it. These are a couple of the pros and cons as I see it:
Pros: It is internally reliable and has been tested lots of times before, and so therefore has a degree of consistency that another (perhaps more directly relevant to climate change scale) would not. The vast majority of the items seem to tap into something fairly timeless and fundamental: the way that humans relate to nature. It is quite difficult to find questions that do that without asking about very time/culture specific things.
Cons: There is something a bit unsatisfying about saying ‘this one is good because its been good before’ but it quite often does come down to that in the business of trying to measure slippery things like attitudes/beliefs etc. I think it is because psychology wants to move closer to the physical sciences in terms of methodology, and so making ‘one tweak at a time’ is seen as good practice (rather than inventing new scales of measurement each time).
I personally do not think that psychology – or at least the kind I am interested in – necessarily benefits from becoming overly obsessed with aping the harder sciences, because you sometimes end up being a bit hamstrung and having less explanatory power than you might otherwise have. I would favour a bit more room for tinkering with scales to make them more relevant to the research question at hand – although that would come at the expense of their proven reliability. I think some of the items are a bit unnecessarily abstract and vague – but then again if they were not, they would quickly go out of date. So what do you make of it…I am guessing you don’t think it’s a useful scale??
I’ve got no quarrel with the concept of measuring attitudes to the environment in this way, as long as we’re clear what we’re measuring.
What struck me about the NEP questions was 1) their complexity and level of abstraction and 2) the very high level of endorsement of the environmental position in the responses to the public opinion survey.
For example: 79% agree that “the balance of nature is very delicate and easily upset”, 74% agree that “the earth is like a spaceship with very limited room and resources” etc. I’d suggest that a large number of people would find concepts like “balance of nature” and “spaceship earth” pretty foreign to their experience; that these people are probably largely represented among the 41% of non-respondents to the postal survey; and that therefore the “don’t know category” should be closer to 50-60% than the 10-20% recorded.
This was a postal survey, and those who were either opposed to environmental ideas, or who found the questions daunting, would be more likely to bin the questionnaire. Add the fact that the survey was conducted among citizens of Washington State, (the “greenest” state of the USA) and you have a couple of strong reasons for doubting if the survey correctly measures public endorsement of the environmental position.
I don’t know if the reason they get high approval ratings is because its an unrepresentatively green sample – I’d say it is because they are phrased in a way that makes disagreeing with them feel a bit odd. You get the same effect when you ask about ‘environmental values’ (this is a different line of research associated with a big body of publications by people like Shalom Schwarz, and Paul Stern) Who really wants to disagree with the ‘value’ of nature? The problem is that people are probably putting very different ideas into that concept of ‘nature’ – and so they then endorse protecting the version in their head.
So yes, the more abstract questions are, the less you can unpick the detail of what people are responding to. Here’s an interesting thought though: before the NEP etc came along – and this goes alongside a turn towards ‘objective’ scales for measuring all sorts of perceptions/beliefs in psychology – the idea that ‘nature’ was something ‘out there’ to be perceived and measured did not have so much intellectual weight.
There is a great book called ‘Contested Natures’ that argues that the clean lines we like to draw between ‘human’ (read: artificial) and ‘natural’ (read: flowers and stuff) are illusory. We shape and are shaped by our environment: a spiderweb is no more ‘natural’ than a guitar. So there are limits – practical and philosophical – to what survey data and questionnaires can do.
This post [here] by a researcher, Stuart Capstick, who is also at Cardiff and did his PhD on reviewing qualitative data on climate change over a long period of time, is interesting, because it suggests that the survey story about attitudes to climate change ‘collapsing’ or ‘rising’ is actually underpinned by some pretty consistent trends.
So I’d like to throw this back to you Geoff – if things like the NEP are flawed, and if you see our scales for measuring climate change scepticism and its relationship to ideology as not capturing the essence of why people are sceptical about climate change, how should social science seek to understand people’s beliefs about the environment and climate change?
To answer your last question first: I’m happy with social scientists asking any questions they like. It’s the quality of the analysis of the answers that bothers me. I find the answers to the NEP questions fascinating, largely because of the large proportion of the sample (even if I’m right that Dunlap’s original Washington State sample might not be entirely representative) who endorse statements that no-one would have thought of formulating sixty years ago. Dunlap and his colleagues are clearly measuring something fascinating - the spread of ideas which have come from nowhere to being majority opinions in roughly half a century.
Take the “Spaceship Earth” concept: Wiki attributes it to the 19th century political thinker Henry George, used by Adlai Stevenson in a speech in 1965 and as the title of a book by Barbara Ward in 1966. I would guess that not 1% of the population have heard of these three people, yet 74% of Dunlap’s sample agree that “the earth is like a spaceship”. This kind of statistic is used, quite naturally, by Greens as evidence of public endorsement of sustainability. Now, if you look at people’s behaviour in our western societies, I’d suggest that “the earth is like a shopping mall” or “the earth is like a crappy theme park” comes nearer to most people’s real opinion. But try proposing those statements in an opinion survey! They correspond to nothing in our conceptual stock.
Dunlap is clearly measuring something important, but what? He and his colleagues clearly believe that they are measuring agreement with a political programme. The NEP has its origins in a book by Dennis Pirages and Paul Ehrlich – “Ark II: Social Response to Environmental Imperatives” (1974) in which it is proposed as an alternative to what they call the Dominant Social Paradigm. A quick Google search suggests that the latter idea hasn’t really caught on, but one understands what it means, and it’s clear that the NEP is expected to replace it in a paradigm shift. Dunlap says that the scale questions “tap primitive beliefs about humanity’s relationship with the Earth”. But he clearly endorses the environmental position as being scientifically attested fact, since he goes on to say:
“we suspect that the never-ending emergence of new scientific evidence concerning the deleterious impacts of human activities on environmental quality and the subsequent threats these pose to humans (and other species) will generate continual pressure for adoption of a more ecological world-view. “
So my concern is not that the NEP is flawed. Nor am I bothered that Dunlap is clearly a committed “green” who believes that the ideas he is exploring as a social scientist are destined to triumph. My concern is that the cultural particularity – the ideological origins if you like – of the concepts used need exploring before the findings of this sort of survey are accepted as neutral objective “fact”.
I’m all for sociological studies of this kind – particularly longitudinal ones which measure changes in public opinion over time – but I believe social scientists are failing in their duty when they fail (or refuse) to interpret data whose meaning is far from clear.
Comments are open for discussion:
Note – ( If need to snip anything as too off topic, I will pass onto Geoff and/or Adam for opinion to see if I’m too heavy handed with the scissors)
Update: added Geoff’s comment to the bulk of the article