Thread by @LoganRStrother

I’m very excited to share some new work, with @SpencerPiston, @EGolberstein, @Sarahgollust, and @DanielEis7, out today in @PNASNews: College roommates have a modest but significant influence on each other’s political ideology pnas.org/content/early/…
1/

Many conservative pundits and politicians have sighted higher education as a front in the culture wars, usually claiming that college education “indoctrinates” students into liberal or leftist ideologies.
2/

We interrogate this claim with an original panel survey, covering the entire freshman class at two universities in the US. We find no evidence of large-scale leftward movement over the course of the first year—most students’ ideology is constant over the 2 waves of the study
3/

With that said, about 23% of students’ ideology changes over that first year. Among those who change, the story seems to be one of moderation:
4/

conditional on changing, students who were conservative at wave 1 tended to move a little to the left, while students who were liberal at wave 1 tended to move a bit to the right.
5/

One interesting point bears mention: because students in the sample skewed liberal at wave 1, the overall sample was slightly more conservative after 1 year of college than it was before. (i.e. There were more liberals who moved right than conservatives who moved left.)
6/

So what explains this movement? There are numerous possible pathways, but we focus on one likely possibility: socialization via roommates.
7/

People tend to be open to socialization when their interest in politics grows, and it is well-established that open discussion of the social world, which is common in college, can induce greater political engagement
8/

In other words, the university environment primes people to be open to socialization.
From here, we take advantage of random assignment of roommates, to see whether assignment to a roommate with different political views influences student ideology
9/

Turns out it does! And the effect tends to be stronger the further apart a student and her roommate are at the beginning of their time together:
10/

In summary, we find 1) no evidence of across-the-board left-ward movement among students in our sample. The data in no way comport to an “indoctrination” story.
11/

2) We find strong evidence that students are influenced by their randomly-assigned roommates; roommates tend to move toward each others’ political ideologies.
12/

The article is open-access (doi.org/10.1073/pnas.2…) and we are interested in your feedback on the piece! We hope this work will spur further study and (maybe!) reduce fears about universities among those on the right, because this is not good for anyone:
13/13