If I read the original blog post correctly, it’s saying that those who identify as Tea Party test higher than the rest of the sample, including liberals. Look at the legend in the 4th plot, it says “teaparty member (n=430)” and “non-teaparty member (n=1886)”. Those numbers add up to the entire data set used in this analysis, not the number of conservatives.
Anyway - a good understanding of basic science & math doesn’t mean they have an unbiased view of economics, or even about science. I used to know an engineer who had a Ph.D in astrophysics and still vehemently denied global warming. We debated a few times and it was clear that he had a good understanding of science, he was just biased in which studies he took seriously. (He also talked about Ayn Rand a lot and had a FairTax sticker, so I think it’s safe to classify him as a Tea Party person.)