Correlation is causation?  Don't be misled by Petr Chylek

It's sort of pesky to be a former teacher of research methods and statistics.  It means that you see huge faults in what is published as science.  Scientists very often don't observe the basic precautions that people such as I have attempted to inculcate in students.  I see it in my own field of social science research, I see it in the medical journals, I see it in climate science journals.

And one of the biggest holes that I see in published research is that the writers ignore just about the first thing you are told in any statistics course:  That correlation is not causation.  Just because two things go together in some way, does not mean one causes the other.  They may both be effects of some underlying third factor or their association might be just a random event.

Let me point out a recent example in climatology: The article "The Atlantic Multidecadal Oscillation as a dominant factor of oceanic influence on climate'  by  Chylek et al.  It is one of the most recent articles appearing in  "Geophysical Research Letters", a major climatology journal.  Here is the Abstract:

"A multiple linear regression analysis of global annual mean near-surface air temperature (1900-2012) using the known radiative forcing and the El Ni¤o-Southern Oscillation index as explanatory variables account for 89% of the observed temperature variance. When the Atlantic Multidecadal Oscillation (AMO) index is added to the set of explanatory variables, the fraction of accounted for temperature variance increases to 94%. The anthropogenic effects account for about two thirds of the post-1975 global warming with one third being due to the positive phase of the AMO. In comparison, the Coupled Models Intercomparison Project Phase 5 (CMIP5) ensemble mean accounts for 87% of the observed global mean temperature variance. Some of the CMIP5 models mimic the AMO-like oscillation by a strong aerosol effect. These models simulate the twentieth century AMO-like cycle with correct timing in each individual simulation. An inverse structural analysis suggests that these models generally overestimate the greenhouse gases-induced warming, which is then compensated by an overestimate of anthropogenic aerosol cooling".

Note the sentence "The anthropogenic effects account for about two thirds of the post-1975 global warming with one third being due to the positive phase of the AMO"

And in the Discussion section of the article we read:  "Our analysis suggests that about two thirds of the late twentieth century warming has been due to anthropogenic influences"

So there you have it:  Global warming has been proven to be mainly  caused by "anthropogenic effects".  When a very sophisticated and careful piece of research comes to that conclusion is there any room left for climate skepticism?

I am afraid there is a very large room left.  The study is correlational:  "A multiple linear regression analysis" and you can't infer causation from correlation.  Yet the sentences I have singled out appear to do exactly that.  An unsophisticated reader would conclude that anthropogenic global warming has now been proven.

Now I feel confident that Chylek and his friends are reasonable people who would be ready to admit to what I have just charged and would say that they were just expresssing themselves in a shorthand way and that they knew from the beginning that the coincidence of temperature rise and CO2 rise in the late 20th century was no proof of anything  -- particularly in the light of the later divergence of those two variables.  But the global warming debate now involves so many people outside the scientific community that I will still charge them with carelessness in the matter.  When unsophisticated people are likely to read your words, you have a duty to make them as clear as you can.

Posted by John J. Ray (M.A.; Ph.D.).