Ok, I'll play (because I like you and find you are usually pretty neutral in regards to Dems Vs Repub bullshit happening).
Lets take one item in the poll I have issue with-
The Democrats and science- Whenever Scientist agree on an issue, we should believe the scientists.
This is the stupidest wording of a poll I've EVER seen, and that's saying something. Of COURSE NOT. Every scientific theory is open to criticism if the evidence flows in a way to be counter to the currently belief. This is TERRIBLE. Now, I'm not saying the conclusion made by the polling isn't unfair, but the wording to get to it isn't exactly doing anyone any favors. There's nothing here to say that a person should't question CURRENT conclusions, or that previous conclusions should't be tested. But of course, it's set up to treat people who say no as "You're a fucking idiot, OF COURSE YOU SHOULD".
Now, do I, as a layman, PRESUME the previous conclusions made by the scientific community, as presented to me, are true and accurate? Uh, YEAH. That's not saying it is, and of course a healthy case of skepticism is important in our day to day lives.
This is just one of the MANY issues I have with the polling done here.
tldr: I agree, but it is no reason to chuck everything out the window, even in this case
You are absolutely right, the questions are not all great. This is true for the Republican questions as well. I don't even think you gave the most egregious example. But the specific questions were not the overall point. It was to test wither over-arching ideals drove people's party affiliation (broad ones, like: Should the government regulate things or should we be accepting of all people). Hundreds of studies use political ideology as the starting point in political affiliation. There is consistent
suggestive evidence that this is probably not correct in the data they collected. This research, again, by an academic that was funded by CAH, is
definitely a pilot study, as he is testing a
new theory. One that flies in the face of usual theories of party affiliation. The results from such research can
only be suggestive and, as such, inform potentially better future studies. The best thing it has going for it is that they recruited a very large sample, by academic standards.
You could probably go through and discard some of the question pairings and it would still leave you with the same or a similar conclusion because there are many opportunities to test how people hold opposing ideals. Remember, that is the aim here. It isn't to try and trick people into looking foolish like some kind of late night show gag (despite how CAH presents the results...don't trust the guys who spent millions of dollars digging a pointless hole for anything). The suggestion supported by these (notably imperfect) results that people stick to their party affiliation in the face of conflicting ideologies is not that outrageous. There is plenty of research in laboratory settings going all the way back to the late 50s showing how people handle conflicting ideas.
I blame the media for it's depiction of science as a series of monumental upheavals. That is
not what we do. We explore and predict to try and figure out what is going on. And a given idea in science will be at one of many different stages of evolution, from suggestive to certain. I think the data these researchers presented was compelling and jives with my experiences (as a moderate). We're still at the "here's a new theory of political affiliation" stage of science but I don't think the sampling method or even the survey design is enough to conclude that the results are biased against one party or another or that they only apply to a small subset of the population. It is possible, though, but it will take a thorough deconstruction of the methods for that case to be compelling. That is what replication and future research is for, of course. One thing we do in science, though, is challenge ideas with data, not just dismiss them out of hand.
(Personally, I've been considering some possible research to testing related ideas using very different methods.)
