Doing good research into any subject is difficult, it is especially difficult on social or economic or similar issues. People always want to know why; did this cause that. It is easy to find spurious connections that are not really proof of any causal relationship. Experimental design has all kinds of traps, the accuracy and completeness of data is always an issue, your choice of analytic and statistical methods will influence the outcome of research. And there are increasing amounts of outright fraud in research. Unfortunately many researchers are ideologues or fame-seekers who will intentionally twist results to show what they want them to show. So you just have to be very, very skeptical about headlines you see in the media regarding supposed research findings.
Here is a great example, from the Ecomodernist substack. Some researchers at Harvard did some research supposedly showing that living near a nuclear power plant was associated with a greater incidence of cancer. The guys at the substack were skeptical, so they looked at the methodology and then replicated it, and replicated it in regard to a variety of organizations at various locations, like a university or a Costco or a major league baseball field. It turns out that the methodology the Harvard researchers used was so flawed that if applied to any location it would find that the location was associated with more incidence of cancer. The researchers fundamental error was not applying their methodology to other types of locations, as the Ecomodernist writers did, and using a very large geographic area to define being exposed to a nuclear power plant.
If it weren’t for independent people like the guys at this substack, we would have no one protecting us from mis-information. The traditional media never check this stuff. The journals that publish some of this dreadful research do a woeful job of peer review or questioning the results. Climate science is full of this crap. I have for a long time advocated for a federal law on use of any research that would require very strict standards before any research could be used to justify a law, a rule or a government action of any kind. The public is entitled to not being misled. (EM Substack)
