Return to Website

Number Watch Web Forum

This forum is about wrong numbers in science, politics and the media. It respects good science and good English.

Number Watch Web Forum
Start a New Topic 
View Entire Thread
Re: March of the repetitions

Hmmmmm, do you suppose he meant 97% of those surveyed?

Re: March of the repetitions

This 97%, so very precise, has been cropping up a lot lately without any reference to the source.
It is a very clever number.
It allows that there are "deniers" but because it is 97% and not 99% or 95% suggests not simply an opinion but an absolute value that results from careful research.

One can almost imagine some activist somewhere in Friends of the Earth or WWF or OXFAM sitting down and saying to him/herself "Now what proportion should I say?" "values like 99% or 95% or even 9 out of 10 sounds too much liek an invented marketing number. Anything less than 95% may suggest there are enough deniers for there to be some genuine vulnerability to the AGW argument. NOt too high as to be unbelievable but 97% sound pretty good and it suggests that it is not 95 or 96% and not an estimate but an actual real value. Yes, let's go with 97%".
ANd of course, 97.4% would be too much like an absolute and people would want to see the research while 97% only suggests it is an absolute value and people are happy to assume it is.

But it is not surprising also to see so many AGW supporters quoting this value with all the certainty that it was presumably designed to ensure.

But does anyone know where this comes from? I mean, the original source. SO much of what we see is actually taken from secondary sources and not the original. It is how they climb the tree from dubious activist websites into more or less assumed to be reliable third party web sites.
It must be fairly recent since I haven't come across it much before this week.

Re: March of the repetitions

Hey, it means it's outside the 95% confidence interval. And it's statistically significant with an alpha of 0.05. Innit?

Re: March of the repetitions

In reply to JMW, I'm slightly surprised that you haven't heard of the "97% of climate scientists agree" claim. It's almost become the Green equivalent of the "8 out of 10 cats" cliche.

The 97% figure wasn't actually cooked up by a Green NGO, it comes from a 2009 paper written by two environmental scientists, Doran and Zimmerman. A fairly good critique of the claim, which also provides a free-to-view link to the paper, is given in this article on the "Free Republic" website:


The D & Z paper is a write-up of an opinion poll survey (personally I don't think opinion polls should be used at all in the hard science literature) and is based on only 79 climate scientists. The sample of 79 is not a random sample (opinion pollsters normally put a lot of effort in to get a random sample) - the survey organisers selected which organisations would participate and 90% were from the USA.

The two questions that were asked were:

1. When compared with pre-1800s levels, do you think that mean global temperatures have generally risen, fallen, or remained relatively constant?

2. Do you think human activity is a significant contributing factor in changing mean global temperatures?

76 out of 79 climate scientists (96.2%) answered 'risen' for question 1, and 75 out of 77 (97.4%) answered 'yes' for the vaguely-worded question 2. For a scientific survey I would have expected question 2 to have made much more of an attempt to quantify what 'significant' means. I can also imagine a lot of climate scientists answering 'yes' to question 2 purely on the basis that if they don't, it looks like their profession is wasting a lot of taxpayers' money.

Often the "97% of climate scientists agree" claim gets hyped into "97% of scientists agree". An example of that is in the recent thread about the World Bank becoming pro-AGW again where the link provided by Francisco includes that claim by the new world bank president.

Re: March of the repetitions

Thanks Dave,
I guess I haven't been paying enough attention. It is just that the figure has cropped up several times recently and I hadn't noticed it before.

Re: March of the repetitions

There was another paper, (sorry no link), which asked authors of recent published papers, (I think they used climate change as a search term,) their thoughts and miraculously came up with exactly the same percentage.
I'll have a look around when I have time.


Re: March of the repetitions

In reply to DaveE, I've got a feeling you're talking about the paper "Expert credibility in climate change" by Anderegg et al. It's available as a free download from this link:

This paper was produced a few months after the Doran & Zimmerman paper, and was partly funded by the Green-leaning 'William and Flora Hewlett Foundation' (William Hewlett was one of the founders of Hewlett Packard). It differs from the D & Z paper in that the '97% consensus' is based on a rather devious definition of 'expert climate scientists' rather than purporting to be based on all climate scientists. The paper claims that 97 to 98% of the top 50 to 200 climate researchers out of a database of 1372 researchers are pro-AGW, with the top researchers being identified on the basis that they are the most prolific contributors to the climate science literature. Relating expertise in climate science to how prolific an author or co-author someone is strikes me as being weird, it's a bit like claiming the giants of the British literary world are people like Barbara Cartland and Enid Blyton (who wrote 722 and 600 books in their lifetimes respectively). A more reasonable definition of an expert climate scientist would I think be somebody who seems to have demonstrated some climate forecasting skill, rather than an ability to churn academic papers out like billy-o.

I suspect that the idea behind the Anderegg paper is to provide a back-up argument for the 97% figure as the D & Z study might be considered a bit 'vulnerable' in the event that somebody else carries out a repeat of the study and comes up with a number somewhat lower than 97%.

The Heartland Institute released a 'policy brief' document a couple of months ago, presumably in case the climate scientist consensus issue happened to come up in the recent US presidential election. This document debunks both the D & Z paper and the Anderegg paper (but slightly modifies the 97% figure to 98%):

Re: March of the repetitions

I suppose it's worth updating this old thread as there was a major development in the 97% consensus saga last month, with the publication of a completely new study which derives the magic 97% number again.

This study was produced by John Cook, who runs the misleadingly named Skeptical Science blog, along with eight co-authors from four English-speaking countries. A free to view copy of the paper "Quantifying the consensus on anthropogenic global warming in the scientific literature"
is available on this link:



We analyze the evolution of the scientific consensus on anthropogenic global warming (AGW) in the peer-reviewed scientific literature, examining 11,944 climate abstracts from 1991–2011 matching the topics 'global climate change' or 'global warming'. We find that 66.4% of abstracts expressed no position on AGW, 32.6% endorsed AGW, 0.7% rejected AGW and 0.3% were uncertain about the cause of global warming. Among abstracts expressing a position on AGW, 97.1% endorsed the consensus position that humans are causing global warming. In a second phase of this study, we invited authors to rate their own papers. Compared to abstract ratings, a smaller percentage of self-rated papers expressed no position on AGW (35.5%). Among self-rated papers expressing a position on AGW, 97.2% endorsed the consensus. For both abstract ratings and authors' self-ratings, the percentage of endorsements among papers expressing a position on AGW marginally increased over time. Our analysis indicates that the number of papers rejecting the consensus on AGW is a vanishingly small proportion of the published research."

It looks like the major difference between this paper and the two previous studies which came up with the 97% figure is that in this latest study the 97% figure relates to those climate scientists who actually expressed an opinion on AGW. Two thirds of the nearly 12000 papers whose abstracts were reviewed were regarded as having taken no position on AGW. I suppose this is an improvement on the study by Oreskes in 2004 (extensively quoted by Al Gore) which also looked at abstracts for papers in the climate science literature and came up with the highly implausible consensus figure of 100%.

My observation earlier in the thread that the 97% figure is "the Green equivalent of 8 out of 10 cats" is even more pertinent following the Cook study. For people not familiar with the "8 out of 10 cats" phrase, which is also the name of a TV comedy panel show on topical news and opinion polls in the UK, it was originally used in a TV advertising campaign by the makers of a cat food called "Whiskas". The advertising slogan in the 1960s was "eight out of ten owners said their cat prefers it", but by the 1980s advertising regulators had forced them to modify it to "eight out of ten owners who expressed a preference said their cat prefers it".

There was a warning last year in 2012 that this Cook study would be coming along, after a hacker/leaker released details about something called "The Consensus Project" that was being discussed in a private forum connected with Cook's Skeptical Science blog.


In addition to carrying out the study the discussions were making a big thing of the idea of 'marketing' the study. One manifestation of this marketing may be that a new blog that has been set up in the Guardian's environment section called "Climate Consensus - The 97%" run by John Abraham and Dana Nuccitelli (Nuccitelli is one of the co-authors of the new paper). This new blog seems to have taken over from Damian Carrington as the spearhead of the Guardian's commentary on AGW.

There has been some criticism of the new study, for example this article from WUWT where some AGW sceptic scientists have complained that their papers were incorrectly classified:


Re: March of the repetitions

96.7% of all statistics are made up. And that's a fact.