Saturday, January 09, 2010

What do scientists think? (last post for awhile)

Many contentious social and political issues are informed by scientific research. Debates on these issues often appeal to the authority of the researchers in this field. Consulting experts is good component of a decision-making strategy when we don't have the time to thoroughly investigate some issue personally.

However, this still requires that we answer two questions:
  1. who are the credible experts?
  2. what do they think?
I'll get back to these issues in a moment, but I first want to point to a underutilized resource for answering these questions: the National Academies, including the National Academy of Sciences(NAS). The National Academies regularly produce reports on issues of interest to the public, including Climate Change, and Evolution. The National Academies assemble committees of experts on a given topic, which then produce reports that state the opinions of the experts and also review much of the evidence behind their opinions. If you want to know what scientists believe about some field of research and why they believe it, first go to the NAS.

Given the availability of these resources, it's distressing that many people harbor severe misimpressions about expert opinion. Typical errors include the misidentification of experts and exaggeration of the influence of fringe opinions within the research community.

One common error in identifying experts is to consider all scientists to be experts in all fields of research, or all members of a field (e.g. biology) to be experts in all subfields within that field. Even within a subfield (e.g. microbial evolution), there are many lines of research being performed and expertise is not evenly distributed. I think that part of the reason for this error is that the public often isn't aware of the vast amount of research being done these days (which is only going to increase as the global economy expands).

It's true that a scientist may be better than a non-scientist at interpreting the debates in a random field of science, but there are still plenty of people who will have greater expertise than that random scientist. This is the idea behind the reports issued by the NAS: non-scientists turn to scientists as a whole for expertise, and scientists then identify those among their ranks with the greatest expertise.

Another common error regarding expert opinion is to identify one or a few experts (or near-experts) with a particular opinion and then interpreting that opinion to be common among the community of researchers. To make it worse, this one dissident's deviation from the standard opinion is often exaggerated and interpreted as dissent from the core components of the standard theory, rather than being dissent from a particular aspect of the theory (I've seen this with evolution, global warming, and HIV as the cause of AIDS).

Dissent is essential to scientific progress, and scientists often give disproportionate attention to dissenting opinions, if for no other reason than that without disagreement there is nothing to research. However, this emphasis on dissent within the scientific community should not translate into giving dissenters excess influence in public discourse. There are many reasons that a person can arrive at an erroneous conclusion, so there is not much information in the fact that people disagree. Going by the principle that there is one correct opinion and many erroneous opinions, we should pay attention when a large number of experts agree on a point, and not give much attention to the scattered and diverse dissensions (unless you are interested in the issue for purely intellectual reasons).

Unfortunately, the public and the press is good at finding a few voices that provide whatever opinion they want to hear, yet are incapable of surveying the opinion of the large number of experts who hold the dominant opinion...probably because those experts are doing research rather than writing blogs or talking to reporters. This is why we have reports from the NAS.

People also feel that they should be directly informed about the issues affecting their lives, and want to see both sides of "the debate", even though the debate is often driven by groups with an economic or ideological interest in persuading others to the view the world in a particular way. For those who want to rely on experts for guidance but not conclusions, the public debate often degrades into an exchange between two experts (e.g. ClimateSci and RealClimate). While the reader will still have nothing resembling expertise, I suppose that they can more directly evaluate the credibility of the two experts. However, they need to keep in mind that if the blogger representing the dominant view does something that discredits his personal opinion, the reader still has to contend with the credibility of everyone else who supports the majority view.