So per this methodology if a scientist didn’t have an opinion as to whether global warming was man made or didn’t believe you could prove it one way or another he was excluded from the sample. In fact 2/3s of the sample were excluded for this reason Accordingly , as has pointed out numerous times, extrapolating this survey into 97% of scientists believe global warming is man made is utter nonsense
Correct.
Correct. If some papers are about other topics, it makes no sense to include them in a discussion about MMGW.
Technically you are right. You can’t say that 97% of all scientists in the world, regardless of their expertise, believe in MMGW. Yes. I agree with that. However, what I think is being asserted here is that of all the research papers concerned with MMGW, 97% of all the research papers on the specific topic of MMGW, contend that GW exists and is anthropogenic. Only 3% disagree and claim that there is not enough evidence to reach that conclusion.
Correct. A paper can be indexed with the keyword ‘global warming’ because it has some relevance to the topic even though it may not be specifically about GW. If it is not about GW, there really is no reason to include it in the survey.
If 97% of the people polled in an exit poll say they voted for candidate A, is it “utter nonsense” to extrapolate that poll to say that 97% of the voters voted for candidate A, even if the exit poll only polled 1% of the voters? Same thing.
No, it is not. The “sampled papers” were all the papers that appeared in a search of papers that addressed some aspect of global warming. Beyond this you also know that the survey was presented as having been done on 12K papers, and not only the third that took a position.
Of course it does if it is implied that the agreement was across the larger group, which is precisely how this study was misrepresented. Here is how it was presented and here is how most people (mis)understand the results:A survey of over 12,000 peer-reviewed climate science papers by our citizen science team at Skeptical Science has found a 97% consensus in the peer-reviewed literature that humans are causing global warming.
And would you not think it deceptive to suggest you had surveyed 13,800 papers and found 97% agreement? This is exactly what has been done here.
This is not accurate. The 12K papers were selected specifically because they were concerned with AGW. These were the search criteria:*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’.
So what happened to the 66% who took no position? Is it not accurate to say that two thirds of all scientists do not agree with the theory of AGW? I didn’t say they disagreed with it but if they have no position they cannot be said to have agreed with it, therefore it is accurate to state they do not accept the theory.
Regardless of sample size, it would be utter deception to say that 97% of voters voted for the Democrat and only 3% for the Republican if 66% of the votes went to the Independent. Unless you’re in Chicago.
The quote you quoted was me. I used the term “sampled papers” and I say what I mean. The definition in my quote was as I have said.
What difference does it make if it was 12K papers or 3800 papers? It is still a heck of a lot of papers. The fact that there were ~8200 papers that did not take a position is irrelevant. When you look at an exit poll that polls 1% of the voters, would you think such a poll was that much more significant if they had polled 3% of the voters? Especially if what they were announcing was that 97% of the voters voted a certain way?
Yes, that is the hole in the Shakespeare papers analogy. If we lumped Shakespeare papers in with global warming papers, it would be wrong to conclude that the authors of the Shakespeare papers thought the same as the authors of climate-related papers. But here we have a group of scientists who are all working in the climate area. The only distinction being that some of them wrote papers taking a position on global warming and some did not. I guess the only way you could salvage your argument would be to claim that only papers agreeing with global warming get published, and therefore the selection process is biased toward that view. But really? Is there any evidence that a large number of scientists wanted to publish anti-AGW papers, but were prevented by hidden forces?
You are right. The 3,800 number should have been featured instead of the 12K number. But would that have made any difference to the impact of the survey? It was silly of them, and unnecessary to the point, for them to emphasize the 12K number so much. But so what?
No, that would be totally inaccurate. You have no information of any kind on the 66% who took no position. Saying that they do not agree with global warming is making an informational statement. It would be accurate to say that you do not know if they agree or not. That is quite a different statement. Saying that they do agree is done only in probability. That is, it is just like an exit poll where you make probabilistic statements about all voters, even though you polled only a few of them.
Correct. This should be easy to understand for anyone who has an acquaintance with the use of keywords to index a paper.
He cannot be correct if he doesn’t hold a degree.
Science by consensus is not science.
You have a mistaken idea about the mechanism by which heat leaves the earth. It does not leave by means of some conduction through the highest layer of air. It leaves by way of infrared radiation from the ground and by visible reflection of sunlight back into space. The height and surface area of the boundary between the atmosphere and the vacuum of space are irrelevant to this process. Besides, even if you doubled the height of the atmosphere you would still not change the area of this interface layer by more than a fraction of a percent. You are clearly barking up the wrong tree here.
Consensus may not be the best way to conduct science, but it is better than any alternative that has been proposed in this thread, which is generally to find somebody who is saying what you want to hear and quote them.
80 years ago, “meteorologists” were getting degrees in the sciences, like physics and chemistry and were able to study weather at universities such as MIT. Many also received training in the military.
Was I speaking to you?
No sure what you mean. The temperatures of ocean waters at different locations and different depths are taken by a variety of measures. It’s **actual **data.