Scientific controversies (even totally artificial ones such as creationism and climate change denial) are outside the scope of what most people, including politicians need to know. They don’t know anything, so why should their opinions matter?
It matters because having an opinion on something you know very little about is almost a perfect job description for a politician. From my time as Senate Intern #16, I saw that 60-70 major issues of national importance crossed the Senator’s desk every week. There’s no way for any human being to express and educated, sensible, let alone correct opinion on these issues, which is why in the real world politicians rely on staffers and outside advisers. Probably the most valuable, and most underrated skill for a politician to have, is knowing which experts to trust, and when to trust them.
While all of this is very true, it’s also true that science is rarely considered a “major issue of national importance.” Even when science is heavily involved, a national security or economic growth lens often seems more apt. At some point us “pro-science” people have to contend with the fact that science just isn’t as central as we want it to be.
Also via Shaping Science Policy, Kate Clancy illustrates bias in science with a case study on menstruation as pathology:
Not only did the idea of the menotoxin become a ubiquitous menace around any reproductively-aged woman, it began to explain pathology. So the menotoxin, which first was an explanation for the presence of menstruation in women, became a way of diagnosing women as ill… and again, since now all reproductively-aged women could secrete it from any bodily fluid at any time, the state of being female essentially made one pathological.
Soon the idea that the menotoxin indicated specific illnesses began to take hold.
“Dr. Schick and I discussed the possibility that the adult female diabetic out of control, the depressed adult female psychotic, and the adult female in the premenstrual phase secreted some common substance in their sweat.” (Reid 1974)
Here, you see premenstrual women compared directly with two pathological conditions: diabetes and psychosis. And all of these relationships, between menstruation and colic, asthma, wilted flowers, are largely observation, case reports, or poorly controlled experiments. When studies do not support the idea of the menotoxin, as with Freeman et al (1934) and two studies cited by Ashley-Montagu (1940) that were not in English, each get dismissed as outliers (even though in Labhardt’s case from Ashley-Montagu, the sweat of men was often as toxic as that of menstruating women).
And this is where I bring it back to my first two points about bias, that science can be biased by the cultural conditioning of those who perform it, and those who tell it. The people who studied the menotoxin really, really wanted to believe in it, to the point that they would ignore negative results and overstate the power of their anecdotes and case studies. The study of the menotoxin spans at least sixty years, maybe ninety depending on which references you consider legitimate, debated in Lancet letters to the editor, and published in several medical journals.