On May 16th of this year, President Obama tweeted this to his (then) 30,000,000+ followers, citing a newly released study on the scientific community's views on global warming:
To no surprise, it received some pretty generous attention in the Twitter-sphere, accumulating over 2,500 retweets and a solid amount of favorites. But, of course, the tweet was just a small glimpse into the widespread media coverage surrounding the research. And in the months following its release, this particular study, including the "97%" statistic, has quickly become somewhat of a poster child for the environmentalist cause.
Now, personally, I don't have a strong, defined stance on global warming. I lean right and hold more skepticism than anything else, but in large part, I never took the time to confirm anything. That might be part of the reason I'm writing this several months after the research was presented.
Nevertheless, I happened to stumble across some seriously misleading information regarding this statistic -- information that almost completely shoots its credibility -- and I couldn't resist laying it out via blog post.
Before going further, take note of this: I realize this is not the only research to supposedly conclude a strong consensus on global warming (we can deconstruct that probably faulty methodology another day). But because of its recent Justin Bieber-esque status among environmentalists, that's gonna be the scope of this post.
The study in question was the one published this year by Australian global warming activist John Cook. Cook analyzed nearly 12,000 abstracts from peer reviewed climate change literature spanning from 1991-2011, selected by a simple keyword search. To be completely fair, Cook did find a 97% consensus... among the scientists that actually expressed an opinion.
Here's the actual text from the actual abstract of the actual study. By the way, AGW = Anthropogenic Global Warming, or human-caused climate change.
So, there you have it! 97% believe climate change is being caused by humans. But just if you count the 4,000 that actually said something about it, and ignore the other 8,000.
Now, it's totally fine & dandy if researchers just want to search for a consensus only among those who offer an opinion.
But that's not what we're hearing. From a consumer perspective, 97% is almost always painted to represent the whole of climate change research.
Obama's tweet was a great example, but it goes far beyond that. Media outlets had a heyday throwing around this number, while giving no indication of the original sample size, let alone any significant revelation on the methodology used. In fact, the link President Obama cited in his tweet is a prime example straight from Reuters.
Did they lie? Technically, no. But were they misleading? With no mention of the original 12,000 abstracts... you betcha.
And from there on out, it's largely been assumed to be true, with the number taking on a cemented, unquestioned role in media culture, as demonstrated by venues like the blog, "Climate Consensus - the 97%," hosted by the Guardian.
And so, Obama's touted it, the media's touted it, and a lot of passionate environmentalists have touted it. But in reality, the number doesn't represent the entirety of the scientific community at all -- it selectively accounts for only a third of it.
Why the other 66% expressed no opinion one way or the other, I don't know. Perhaps they found the data to be inconclusive, or maybe they wanted to avoid immersing themselves in any political fuss. Regardless, if you do want to represent the work of all scientists, HERE is a far more accurate, honest way to explain the data:
Based on an analysis of 12,000 abstracts, less than a THIRD of all climate change scientists expressed the belief that global warming is man-made.
Why not phrase it that way? Probable possibility: It would set back an important, widespread agenda that's been rapidly accelerating the past several years. It's far easier, less damaging to reputations, and cheaper to pay attention to a select 4,000, and come to a more convenient conclusion because of it.
It's dishonest not only to the scientific community, but to us.
And we're not even done yet.
If you dig further into some of this methodology, you'll be even more enlightened.
The language President Obama used in that go-to tweet was pretty strong: "#climate change is real, man-made and dangerous." In reading it, you kind of get the vibe that these scientists didn't nonchalantly mention their opinion to Barack at a White House tea party. Rather, it's a stronger, more confident, urgent claim that has serious implications for you & me.
... But that's not exactly what the research indicated either.
As a means for measuring the opinions represented in the research, Cook designated a seven-level system for evaluating the abstracts' views on global warming. As you can see, the top three were classified as "endorsements," while the bottom three were "rejections." (Screenshot of the table found in the actual study.)
And below is a breakdown of where the abstracts fell into place. (Note: You can manually find the number of abstracts in each respective category here, but I took the word of Door Marcel Crok's own number crunching to save time.)
Category 1: 65
Category 2: 934
Category 3: 2933
Category 4: 8261
Category 5: 53
Category 6: 15
Category 7: 10
Total: 12271
No, that's not a typo. There actually were only 65 abstracts assigned to the top, most confident (quantified) level. Out of the 4,000 that expressed an opinion (which would exclude category 4), that's about 1.6%. Out of the original sample of 12,000, it's about .5%.
Read that again: one half of one percent of the entire 12,000 abstract sample confidently and quantitatively said, "Global warming is definitely caused by humans."
And even if you group together the explicitly endorsed abstracts (categories 1 & 2), you still get a weak 25% of expressed opinions, and just 8% out of the entire sample.
... doesn't sound so hands-down to me.
Today's lesson: Don't blindly swallow over-sensationalized statistics. Especially when you hear them from Barack Obama.
In other, totally unrelated news: There's a new book on my reading list. It's called How to Lie with Statistics by Darrel Huff and Irving Geis. If you were as appalled by this as I was, I recommend it to you as well.
Ninety-seven percent of scientists agree: #climate change is real, man-made and dangerous. Read more: http://t.co/4lEEBYtVqf
— Barack Obama (@BarackObama) May 16, 2013
To no surprise, it received some pretty generous attention in the Twitter-sphere, accumulating over 2,500 retweets and a solid amount of favorites. But, of course, the tweet was just a small glimpse into the widespread media coverage surrounding the research. And in the months following its release, this particular study, including the "97%" statistic, has quickly become somewhat of a poster child for the environmentalist cause.
Now, personally, I don't have a strong, defined stance on global warming. I lean right and hold more skepticism than anything else, but in large part, I never took the time to confirm anything. That might be part of the reason I'm writing this several months after the research was presented.
Nevertheless, I happened to stumble across some seriously misleading information regarding this statistic -- information that almost completely shoots its credibility -- and I couldn't resist laying it out via blog post.
Before going further, take note of this: I realize this is not the only research to supposedly conclude a strong consensus on global warming (we can deconstruct that probably faulty methodology another day). But because of its recent Justin Bieber-esque status among environmentalists, that's gonna be the scope of this post.
The study in question was the one published this year by Australian global warming activist John Cook. Cook analyzed nearly 12,000 abstracts from peer reviewed climate change literature spanning from 1991-2011, selected by a simple keyword search. To be completely fair, Cook did find a 97% consensus... among the scientists that actually expressed an opinion.
Here's the actual text from the actual abstract of the actual study. By the way, AGW = Anthropogenic Global Warming, or human-caused climate change.
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. [Emphasis my own.]
So, there you have it! 97% believe climate change is being caused by humans. But just if you count the 4,000 that actually said something about it, and ignore the other 8,000.
Now, it's totally fine & dandy if researchers just want to search for a consensus only among those who offer an opinion.
But that's not what we're hearing. From a consumer perspective, 97% is almost always painted to represent the whole of climate change research.
Obama's tweet was a great example, but it goes far beyond that. Media outlets had a heyday throwing around this number, while giving no indication of the original sample size, let alone any significant revelation on the methodology used. In fact, the link President Obama cited in his tweet is a prime example straight from Reuters.
Experts in Australia, the United States, Britain and Canada studied 4,000 summaries of peer-reviewed papers in journals giving a view about climate change since the early 1990s and found that 97 percent said it was mainly caused by humans.
Did they lie? Technically, no. But were they misleading? With no mention of the original 12,000 abstracts... you betcha.
And from there on out, it's largely been assumed to be true, with the number taking on a cemented, unquestioned role in media culture, as demonstrated by venues like the blog, "Climate Consensus - the 97%," hosted by the Guardian.
And so, Obama's touted it, the media's touted it, and a lot of passionate environmentalists have touted it. But in reality, the number doesn't represent the entirety of the scientific community at all -- it selectively accounts for only a third of it.
Why the other 66% expressed no opinion one way or the other, I don't know. Perhaps they found the data to be inconclusive, or maybe they wanted to avoid immersing themselves in any political fuss. Regardless, if you do want to represent the work of all scientists, HERE is a far more accurate, honest way to explain the data:
Based on an analysis of 12,000 abstracts, less than a THIRD of all climate change scientists expressed the belief that global warming is man-made.
Why not phrase it that way? Probable possibility: It would set back an important, widespread agenda that's been rapidly accelerating the past several years. It's far easier, less damaging to reputations, and cheaper to pay attention to a select 4,000, and come to a more convenient conclusion because of it.
It's dishonest not only to the scientific community, but to us.
And we're not even done yet.
If you dig further into some of this methodology, you'll be even more enlightened.
The language President Obama used in that go-to tweet was pretty strong: "#climate change is real, man-made and dangerous." In reading it, you kind of get the vibe that these scientists didn't nonchalantly mention their opinion to Barack at a White House tea party. Rather, it's a stronger, more confident, urgent claim that has serious implications for you & me.
... But that's not exactly what the research indicated either.
As a means for measuring the opinions represented in the research, Cook designated a seven-level system for evaluating the abstracts' views on global warming. As you can see, the top three were classified as "endorsements," while the bottom three were "rejections." (Screenshot of the table found in the actual study.)
And below is a breakdown of where the abstracts fell into place. (Note: You can manually find the number of abstracts in each respective category here, but I took the word of Door Marcel Crok's own number crunching to save time.)
Category 1: 65
Category 2: 934
Category 3: 2933
Category 4: 8261
Category 5: 53
Category 6: 15
Category 7: 10
Total: 12271
No, that's not a typo. There actually were only 65 abstracts assigned to the top, most confident (quantified) level. Out of the 4,000 that expressed an opinion (which would exclude category 4), that's about 1.6%. Out of the original sample of 12,000, it's about .5%.
Read that again: one half of one percent of the entire 12,000 abstract sample confidently and quantitatively said, "Global warming is definitely caused by humans."
And even if you group together the explicitly endorsed abstracts (categories 1 & 2), you still get a weak 25% of expressed opinions, and just 8% out of the entire sample.
... doesn't sound so hands-down to me.
Today's lesson: Don't blindly swallow over-sensationalized statistics. Especially when you hear them from Barack Obama.
In other, totally unrelated news: There's a new book on my reading list. It's called How to Lie with Statistics by Darrel Huff and Irving Geis. If you were as appalled by this as I was, I recommend it to you as well.