Professor Logan Strother of Purdue University has responded to my critique of his study on higher education’s effects on student ideology. I’ll take his objections point-by-point.

1. Strother says that “Schorr thinks we would also not find evidence of indoctrination if we looked at the full body of students over the course of their entire college careers. Rather, he thinks this happens among a v. small (but esp important?) subset of college students.” No, not exactly. It’s facially unlikely that the general ideological shift of students over the course of four years on campus would not be to the left, and it would belie the experience of most students — left, right, and center. My point is only that since most students aren’t especially concerned with politics and have no intention of pursuing a career related to it, I would expect the shift to be less dramatic among the general population than among the politically interested. And yes, of course the ideological shift of those who want to go into politics, law, journalism etc. is especially important.
2. Strother also says that my “argument that ‘peer pressure’ and ‘incentives for radicalization’ among ‘the politically interested’ are the key driver of ‘a leftward shift in students’ politics’ is interesting” but one that is “entirely outside the scope of our study, which is primarily about roommate influence on ideology.” He can’t have it both ways then. If the study is only about the effects of roommates over the course of their freshman year, then fine, that’s interesting enough! But he can’t use his study to “disprove” conservative complaints about the ideological effects of the four-year college experience, and then retreat to “it’s entirely outside the scope of our study” when challenged.
3. Strother asks me if I have evidence to prove my claims about the way “indoctrination” (his word, not mine) works on campuses. There is no hard data on this, and it’s also an admittedly difficult thing to “prove” through a study. For students on the ground, however, there can be no doubt of this phenomena. At my own alma mater of Cornell University, the presidents of the Student Assembly and secret society Quill and Dagger spent the last semester attempting to not only disarm the university police department, but persecuting the members of the Assembly who opposed their efforts to do so. During my own time on campus, the editor-in-chief of the student newspaper (now a blue-checkmark journalist) declared proudly that he considered it his collegiate mission to destroy the College Republicans. Should this be surprising? Of course not on a campus where 98-99 percent of staff and faculty donate to Democratic candidates and left-wing causes. A cursory glance at the campus scene around the country will reveal dozens if not hundreds of incidents similar to or worse than these. I’m sure the retort will be “the plural of anecdote is not data,” but the truth is that much of the political goings-on at college are hard to pick up in studies. 
4. Strother defends his use of self-identification as a measure of students’ ideology, calling it “quite important” and citing a study published in the Annual Review of Psychology back in 2009. Ideological self-identification is no doubt “important,” but it’s certainly not a better measure of one’s views than asking them a range of philosophical and policy-oriented questions. He also asks why conservatives and moderates don’t move to the right and far-right. Well there is a slight shift into the conservative category, as he himself acknowledges. I would suspect that there aren’t more far-right respondents because few people like to think of themselves as “far-anything” much less far-right, which is associated with the detestable alt-right and even neo-Nazis.

Strother is clearly motivated to prove that the college experience doesn’t move students to the left ideologically, but his study is poorly suited to do that, and it’s silly for him to avow that it does.