You may have seen a link to this posting by M.S. on The Economist blog, which has drawn some blogospheric attention. Only 6% of scientists identify themselves as Republicans, compared to 55% who identify as Democrats. M.S. wonders why:
I can think of three testable hypotheses they might look into. The first is that scientists are hostile towards Republicans, which scares young Republicans away from careers in science. The second is that Republicans are hostile towards science, and don't want to go into careers in science. The third is that young people who go into the sciences tend to end up becoming Democrats, due to factors inherent in the practice of science or to peer-group identification with other scientists. In the absence of data, I leave it to you to decide which you find most plausible. But by all means, social scientists should look into this.
Hey, I'm a social scientist! But I don't have to look into it, I already know the answer, and it's None of the Above.
Scientists study reality, carefully applying rigorous methods of data collection and rules of inference. They think from evidence to conclusions. I don't need to tell you that reality has a well-known liberal bias. That's why conservatives don't go near it. They reason from a priori principles to decide that reality must be whatever they need it to be in order to believe what they already believe, which isn't actually true. So they can't be scientists.