Statistics for Language in Teaching Evaluations

[beero1000]

http://benschmidt.org/profGender/#

A visualization of language in 14 million ratemyprofessor.com evaluations, specifically analyzed by field, professor gender and positivity.

Really interesting stuff. Some (unsurprising?) results I've found with respect to math (my field): "stupid", "bad at", "too fast", "too slow", "explained".

Also check out some of the gender/field differences: "partial credit", "arrogant", "humble", "unfriendly", "too smart", "genius", "corny", "funny", "humor", "handsome", "unqualified", "sexist", "racist", "jerk"

Post if you find a good one.

 

[Kharakov]

In your field (15-20 years ago!), a lot of the TAs who taught the "labs" (not the professors) were hard to understand because they have thick accents.

Looked at the data.. it appears that there are still language/accent barriers to overcome in physics/math/economics.

 

[ronburgundy]

Somewhat surprising that many appearance words occur more for male professors:

hot, hott, cute, sexy, hair, wears


I wish you could break the ratings down by the gender of the student too. In fact, some of the difference for male and female profs may have more to do with how male and female students differ in the what they focus on for same vs. opposite sex profs.

 

[ronburgundy]

Not surprising, is the use of the word "liberal", which doesn't differ by gender but predictably by discipline with the softest sciences and Humanities having the most mentions, then the more empirically grounded social sciences, like economics, psychology and health sciences, with the hard sciences and math-based fields having almost no mentions.

It isn't that the hard scientists and math folks are "conservative" (that word is rarely mentioned), but just less aggressively liberal to the point where their stance is obvious to students.

 

[Deepak]

Coming off a redeye and a full day of work so I haven't had much time to explore, but the 'funny' results seem to back up Hitchens...

 

[ronburgundy]

Coming off a redeye and a full day of work so I haven't had much time to explore, but the 'funny' results seem to back up Hitchens...

Possibly, though female profs are still on average younger and less likely to have tenure, both of which could feed into an effort to be more serious, and less relaxed, especially when teaching college students only a few years younger than you and ready to pounce on any basis to disrespect you.

 

[Peez]

can you find an adjective that was never used?

I found that "mastodon" was never used, but I could not find an adjective that would normally be used to describe a person and that was never used in the data.

Peez

 

[ronburgundy]

I found that "mastodon" was never used, but I could not find an adjective that would normally be used to describe a person and that was never used in the data.

Peez

Well, there are about 50 words per review, and as many as 750,000 reviews per subject (the highest for English), which is up to 37,500,000 words for that subject, and maybe around 10,000,000 words for the average subject. Probably few adjectives would fail to be used at least once in 10 million words of description.

Even "bawdy" and "resplendent" has a couple of mentions.

That does raise the point that unless its something mentioned at least 10 times per million words, then any gender of subject differences don't mean much of anything.

 

[JonA]

Why isn't this in the Social Sciences forum?