The archenemy of scientific thinking is conversation. As in typical human conversational discourse, much of which is BS. Personally I have become rather fed up with talking to people. Seriously, it is something a problem. Fact is, folks are prone to getting pet opinions into their heads and then actually thinking they are true to the point of obstinacy, even when they have little or no idea of what they are talking about in the first place. We all do it. It is part of how the sloppy mind generating piece of meat between our ears we call the human brain is prone to work. Humans may be the most rational beings on the planet these days — but that's not saying much considering that the next most rational are chimpanzees.
Take creationism. Along with the global climate issue and parental fear of vaccination, the fact that a big chunk of the American body politic deny evolutionary and paleontological science and actually think a god created humans in near historical times is causing scientists to wonder just what is wrong with the thinking of so many people — mass creationism has been used as a classic example of mass anti-scientific thinking by others responding to this question. But I am not going to focus so much on the usual problem of why creationism is popular, but more on what many who promote science over creationism think they know about those who deny the reality of Darwin's theory.
A few years back an anti-creationist documentary came out, A Flock of Dodos. Nicely done in many regards, it scored some points against the anti-evolution crowd, and when it came to trying to explain why many Americans are repelled by evolution was way off base. The reason it was so wrong was because the creator of the film, Randy Olson, went to the wrong people to find out where the problem lies. A (seeming) highlight of the picture featured a bunch of poker playing Harvard evolutionary scientists gathered around a table to converse and opine on why the yahoos don't like the results of their research. This was a very bad mistake for the simple reason that evolutionary scientists are truly knowledgeable only about their area of expertise, evolutionary science.
If you really want to know why regular folk think the way they do then you go to the experts on that subject, sociologists. Because A Flock of Dodos never does that, its viewers never find out why creationism thrives in the age of science, and what needs to be done to tame the pseudoscientific beast.
This is not an idle problem. In the last decade big strides have been made in understanding the psychosociology of popular creationism — basically, it flourishes only in seriously dysfunctional societies, and the one sure why to suppress the errant belief is to run countries well enough that the religion creationism depends upon withers to minority status, dragging creationism down with it.
In other words better societies result in mass acceptance of evolution. Yet getting the word out is proving disturbingly difficult. So the chatty pet theories abut why creationism is a problem and what to do about it continue to dominant the national conversation, and pro-creationist opinion remains rock steady (although those who favor evolution without a God is rising along with the general increase of nonbelievers).
It's not just evolution. A classic example of conversational thinking by a scientist causing trouble was Linus Pauling's obsession with vitamin C. Many ordinary citizens are skeptical of scientists in general. When researchers offer up poorly sustained opinions on matters outside their firm knowledge base it does not help the general situation.
So what can be done? In principle it is simple enough. Scientists should be scientists. We should know better than to cough up committed but dubious opinion on subjects outside our expertise. This does not mean a given scientist has to limit their observations solely to their official field of research. Say a scientist is also a self-taught authority on baseball. By all means ardently discuss that subject the way Stephen Gould used to.
I have long had an intense interest in the myths of World War II, and can offer an excellent discourse on why the atom bombing of Hiroshima and Nagasaki had pretty much nothing to do with ending the war in case you are interested (it was the Soviet attack on Japan that forced Hirohito to surrender to save his war criminal's neck and keep Japan from being split into occupation zones like Germany and Korea). But if a scientist finds him or herself being asked about something they do not know a lot about either decline to opine, or qualify the observations by stating that the opinion is tentative and nonexpert.
In practical terms the problem is, of course, that scientists are human beings like everyone else. So I am not holding my breath waiting for us to achieve a level of factual discourse that will spread enlightenment to the masses. It's too bad but very human. I have tried to cut down on throwing out idle commentary without qualifying its questionable reality, while being ardent about my statements only when I know I can back them up. Me thinks I am fairly successful in this endeavor, and it does seem to keep me out of trouble.