Doran & Zimmerman, Anderegg, Lewandowsky
These were all filtered polls with lousy methodology. None of them would hold up on their own, but the disingenuous will group them to make it look like there's more evidence than there is.
- 2009 Doran & Zimmermann They polled 3,146 Government Earth Scientists (none from private sector), then filtered all but 77 that weren't published in a few pro-Global Warming journals, asked them two vague questions and concluded that 75 of the 77 were pro-AGW (Anthropogenic Global Warming). His question standards were far below the IPCC's "most" or ">90%" of Global Warming is caused by man. It was sloppy conclusions so activists were happy to move to the Cook "Study", because this one had little credibility, but it still had more than Anderegg.
- 2010 Anderegg had worse methodology than Doran & Zimmerman. This one filtered to 908 based on their ability to publish more than 20 papers in pro-AGW publications, and concluded that 97% of those papers didn’t disagree that man was "a major contributor" to current warming. Hardly meeting the IPCC standards or conclusive of anything. Most people didn't cite Anderegg.
- 2010 Lewandowsky used other datasets (Oreskes, Anderegg and Doran) which had overlapping data (thus over-counted). Since your conclusions are only as good as your sources, and all the sources were discredited, we say GIGO (Garbage-in, Garbage-out), and virtually no one cites Lewandowsky, other than some like Cook who throw it in to their study of studies to bias the results.
2009 Doran & Zimmerman
This was the most common source for 97% myth, up until the Cook "Study". Prof. Doran's had his graduate student (Margaret R. K. Zimmerman), poll 10,257 Earth Scientists at academic and government institutions, asking two questions, and they got 3,146 that responded. Her paper on it, "The consensus on the consensus”, is less often cited, since Doran invented the more sensational conclusions, and has the bigger pedigree.
Doran then threw out 3,069 of the responses because they were in adjoining sciences (and weren’t published in a Climatology Journal), and focused on the remaining 77. And shockingly, 75 of the 77 that they decided to count, agreed with both questions. And that’s where Doran got the public claim that 97% of scientists agreed that global warming was man made (by ignoring 10,182 responses). More than that, if you look at the 3,069 people disqualified, we find that 50% of Geologists, 38% of Meteorologists and 20% of all Climatologists don’t believe man was a significant contributing factor to Global Warming. (Remember Doran threw everyone out but Climatologists that published in a few pro-AGW Climatology Journals, if they were published elsewhere, they didn’t count). Geologists and Meteorologists also study climate (in fact, Climatologists get most of their data FROM the geologists and meteorologists who do the ice core, sediment and temperature observations), but since Geologists/Meteorologists grants don't come from providing politicians with political fodder, only 50-62% of them think global warming is "significantly contributed" by man, or this warming is unusual. So ironically, the Doran study is used by both sides, depending on if you look at the entire data-set, or the doctored subset that shows what the Climate-scarists want.
The two qualifying questions
- When compared with pre-1800s levels, do you think that mean global temperatures have generally risen, fallen, or remained relatively constant?
- Do you think human activity is a significant contributing factor in changing mean global temperatures?
Of course saying that man might be a "significant contributing factor" to warming is not close to the IPCC AR4 "very likely" (90% certainty), "most warming" (70-90%) claims, let along the IPCC AR5 stronger claims. Almost everyone agreed that man’s 14-21B tons of CO2 to the 22,056,773B tons in the system (773B cycling yearly) is probably contributing a little to that warming, but that’s well below most or >90%. And it’s certainly not agreeing with any policy decisions about how we should respond. With 44% of the total agreeing man is even a significant contributor, that hints strongly that with the U.S. only being responsible for 12% of all CO2 contributions by man, that Kyoto or other radical responses would not be the right solutions.
- ≈56% of all scientists we sampled (assuming even distribution of Geologists/Meteorologists) agree with a soft statement like “was Global Warming contributed by man”. With Geologists and Meteorologists much more skeptical than folks who get their paycheck from Climate-Scarist funding
- But by the Press/Doran reasoning, “97% of the (75 of the 77 people we cherry picked to agree with us), agreed with us”, if you ignore the 3,069 that responded that we ignored.
This was similar to Doran. But it’s less quoted as it was even less valid. It was a “study" of 1,372 researchers (defined as top publishers to place that were later demonstrated to be excluding dissenters in Climate-Gate). Then filtered down to 908 people by qualifying only people that authored 20 climate publications or more. Then they concluded that 97% didn’t disagree that man was "a major contributor" to the warming in the 20th century. Not disagreeing isn’t necessarily agreeing (since it includes no opinion or people that skip the question), and this is well below IPCC AR4 "very likely" (90% certainty), "most warming" (70-90%) claims. And it was such a filtered subset as to be statistically useless.
Many parts of the “study” were pilloried as too subjective to be useful. So it isn’t often used. People went from citing Doran to citing Cook, so this is a non-player.
This just re-used the other studies datasets (Oreskes, Anderegg and Doran), then tried to use an aggregate study to re-prove the same things. But in computers and statistics we say GIGO (Garbage-in, Garbage-out): you’re only as good as your source data. Since all the other sources data had been discredited (and they did the root research), so was this study. In fact, since many of the papers would be overlapped in the studies, it was multiple counting and thus less credible than any of the individual studies. Thus everyone pretty much ignored this regurgitation.