How to properly check for conspiracy ideation in climate change science believers and skeptics
To adequately address the correlation (but not causality) of skeptics and conspiracy theory:
- A large sample size (preferably 500 or more—the larger the sample, the more reliable the results and the higher the probability of it being truly random). If one wants to use the internet, a Google ad on a wide variety of websites should provide a reasonably random sample. As a subset, a Google ad targeting climate change site searches (both sides) could assure the taker is familiar with the issue. The wider sample would apply to the general public. All must be meticulously documented—what sites the ads were posted on, what ads and sites had hits and surveys filled out, etc, in case there are questions about the methodology or sampling procedures.
Ideally, one would use a mass mailing. However, this would be quite expensive (if one figures only about 10% or less of the survey recipients will return filled out survey) and would only be practical if funding is available.
If using an internet survey, care must be taken to make sure one individual does not repeatedly fill out the survey. This could be done with the use of a password unique to the survey taker or a check of the IP address to verify the same computer is not responding. (This is not foolproof. One eliminates as much of the problem as possible and lives with the rest.)
- A survey that avoids as much prejudice as possible. It should include a title that is neutral (such as “Social Science Study”.
Questions should be arrange randomly on the survey, meaning there are multiple versions of the survey that differ only in the order of the questions. This helps assure that the order of the questions do not influence the answers.
As for the questions:
One may ask political party affiliation (for later use when you want to use the survey to correlate politics and climate change), random information (college degree, occupation, if one believes in big government, if one gives to charities, etc). The random questions yield information but also draw undue attention from the point of the survey, helping insure the answers to the climate and conspiracy theories are not unduly influenced.
Next, include the questions on conspiracy and science theories. The wording of these questions must be as neutral as possible. For example, “Do you believe smoking causes cancer?” is not suitable. “Do you believe smoking causes cancer in some people?” People who are very precise with language may answer “no” to the first version because they know people who smoked and did not get cancer. Another variant is “Do you believe studies that show a link between smoking and cancer?” Questions like these and “Do you believe vaccines cause autism?” measure how much one believes in the accuracy of scientific research. It indirectly may indicate one’s belief in science itself. More properly, it indicates the level of skepticism the individual has in some areas.
For conspiracy theory, one asks about 9/11, faked moon landing, illuminati, UFO hidden research (not UFO’s in general—there are unidentified objects flying around), aliens, one world government, etc.
When addressing climate change, a series of questions is preferable. Does the person believe the climate is changing more than in the past? How much of this does the person attribute to humans? (A sliding scale of 1 to 10 or something similar will give a more accurate reading.) How much does the person trust the IPCC to accurately report climate change? Is consensus important in science? How far should humans go in trying to stop climate change (list of proposals with y/n options—things like switching to renewables, shutting down coal plants, giving money to the UN, and so forth. The proposals should range from minor changes to worldwide changes, voluntary versus involuntary).
If one does not get a wide variety of responses from both sides of the debate, the ad should be ran again. Ideally, the pro and con should be close to equal. If you cannot achieve this, it needs to be clearly indicated in the study.
The data from the surveys is then compiled. One then separated belief from skepticism (if you used a sliding scale for belief that humans caused climate change, this variable should be subdivided to each section of the scale—how many believed it was a 5, how many believed it was a 6, etc). One then separates the conspiracy/science belief data for each group. Then one calculates what percentage of each group answered affirmatively to the conspiracy/science questions. Both sides scores must be reported. It is not appropriate to report only one side.
This shows only the correlation between beliefs, not the reason or cause for the beliefs. That should be made clear. It may be that the people who believe in conspiracies are more likely to question things, resulting in a higher number questioning or disbelieving climate change. This does not mean that those who question climate change are conspiracy theorists—it means conspiracy theorists question things more and are less trusting, resulting in their questioning climate change. They may perceive it as a conspiracy, also. Not all persons will believe in climate change and conspiracies, not all skeptics will believe climate change is in error and conspiracies are true.
Such a study is only useful if one is looking for ways to approach people about climate change science in the hope of changing their views to reflect the view preferred by the presenter. If one wants to know how to convince skeptics of climate change and human causes, it may be useful to know if they are also believers in other parts of science. It does not prove the individual’s belief in climate science is wrong—people can easily be wrong in one area and right in another.
If you use the internet, you only survey those who are into electric media. This may not be a serious limitation since much of climate science is debated on blogs, but it is worth noting. Persons who are very much into conspiracy theories may avoid the internet. Their beliefs will not be represented. A mailing might help, though admittedly some persons will be too suspicious to answer the survey.
Results of the survey, if properly set up, can be used to look at other correlations between religion, politics, etc. Again, none of this addresses the truth or falsity of the beliefs, only correlation. The results may be useful in formulating a possible way to convert those with unwanted views (unwanted from whomever’s point of view) or to attempt to discredit the opposing view, as in the case of the unscientific Lewandosky. In other words, this is a marketing study.