Thread by @JeffreyASachs

The Myth of Partisan Grading!

There's been a lot of talk lately about self-censorship on campus. According to one popular theory, conservative students censor themselves because they believe that if they state their true views, they will get a lower grade.

This theory is true.

By which I mean that it IS the case that conservative students believe this. I know offhand of two surveys confirming it. A 2017 @TheFIREorg survey shows conservatives report self-censoring in the classroom more than liberals, especially at the extreme ends of the spectrum.

And when asked why, conservatives are more likely to say they have self-censored due to fear of receiving a lower grade from their professor.

Source [Tables 12 and 13]: thefire.org/publications/s…

Similar findings were reported in a 2017 @HdxAcademy survey, which added a question about which TOPICS generate self-censorship. Again, conservatives were more likely to self-censor due to fear of a bad grade.

heterodoxacademy.org/the-fearless-s…

OK, what does this mean? Well, it means that there's a significant percentage of conservative students (about 10%, according to FIRE) who believe that if they express their genuine opinions in class, they will be penalized by their professors.

But guess what? They're wrong.

We have a couple studies looking at this exact issue. An experiment by Musgrave and Rom (2015) had polisci students compose two essays, one on the history and beliefs of the Democratic Party and another on the GOP. These were then given to TAs to grade.

scholar.dickinson.edu/faculty_public…

The result? Neither the partisan affiliation of the students nor of the TAs, or the topic of the essay, had any impact on the grades awarded. In other words, no bias.

When presented with these findings, many anti-Left activists react with incredulity. See e.g. here, where James Lindsay of Sokal2 fame likens it to climate change denialism. Ironic, considering he's the one using anecdote to refute empirical research.

https://twitter.com/ConceptualJames/status/1095087617299894283

"But wait!" you cry. "That's just political science. The REAL problem is in the so-called "grievance studies" disciplines (gender studies, anthro., critical race studies, etc.). If any part of the academy awards conservatives lower marks than liberals, it's there, right?"

Nope.

In Kemmelmeier, et. al. (2005), researchers found that even after controlling for student ability (via SAT scores), conservatives received higher grades than liberals in "Hierarchy Enforcing" (HE) disciplines like economics or marketing.

wolfweb.unr.edu/homepage/marku…

However, conservatism had NO impact on grades in Hierarchy Attenuating (HA) fields like American culture, African American studies, cultural anthropology, education, nursing, sociology, and women’s studies. Again, this is AFTER controlling for student ability.

In other words (hold on to your butts), the more "woke" the discipline, the more fair the grading. Or to quote the authors:

wut? is? going? on?

OK, freakout over? Ready to keep going? Because we're not done yet.

While the "liberal prof gives F to conservative student" theory is a myth, this isn't to say that partisanship doesn't have any effect on grading. It does. Just not in the way you might think.

One fascinating paper (Bar and Zussman 2012) finds that Democratic profs are more "egalitarian" in their grading while Republican profs award more grades at the higher/lower extremes. Republicans also systematically give lower grades to black students.

talia-bar.uconn.edu/wp-content/upl…

Note that there are many possible explanations for the racial disparity besides discrimination (whether "positive" in the case of Dems or "negative" in the case of Reps). And note also that BOTH kinds of profs systematically give black students lower grades. Dems just do it less.

But it's not due to differences in student ability. That again gets controlled for via SAT score matching.

(The report finds little partisan difference in grading for female vs. male students)

OKAY. What's the tl;dr of it all?

First, to the liberals in the audience: I bet you're feeling pretty smug about now, right? Well, stop it. Yes, the conservatives-are-penalized theory is a myth, but that doesn't mean they don't face other challenges on campus.

As I said at the top of the thread, they are more likely than libs to self-censor. They are also less likely to be invited to socialize with profs out of class or be asked to collaborate on research projects. Lots of other ways. It can be rough out there. Faculty must do better.

Now to the conservatives in the audience: You probably think I'm some kind of leftwing hack. You're right! But I also tried to be as fair and transparent in this thread as possible, supplying you with all the citations and ungated sources.

One response you might have: If there's no evidence of anti-con grading bias, it's only because cons have by necessity gotten really good at parroting back to liberal profs whatever they want to hear. Checkmate, Sachs!

I don't think so. The overwhelming majority of cons feel comfortable expressing themselves in class (78%), though this is lower than the percentage of liberals (89%). And most reasons given for self-censorship have nothing to do with the prof.

FIRE survey, Tables 9 and 13.

And recall that in the Musgrave and Rom article, the essays were part of a controlled experiment (i.e. the grades didn't count), so conservative students had no incentive to pretend to have more liberal opinions than they did in reality.

Lastly, you might fire back by recalling the time you personally were docked points for voicing a conservative opinion. I believe you! Raging SJW that I am, I take your auto-ethnographic lived experience standpoint epistemology seriously.

But it's not enough. And if you're really the facts-over-feelings skeptic you believe yourself to be, this may be the moment to do that one thing we're all trained from birth to never ever do: change your mind.

EDIT: sorry, more like 20%. My mistake.