Sure, your average philosophy grad student is not as rational as your average LW meetup attendant or your average MIT student, but the evidence does suggest that they are more rational than your average grad student (assuming the CRT is a reasonably reliable test of rationality, of course). Surely that says something about the benefits of philosophy training. There are pretty strong selection effects associated with attending LW meetups and going to MIT. If that’s your bar for passing the CRT, it’s a very high bar.
I’d also note that looking at all people with some graduate training in philosophy is casting a pretty wide net. There are a lot of pretty bad philosophy graduate programs out there. I’d be interested in seeing a comparison of, say, students at the top 20 philosophy graduate programs with the groups mentioned in Luke’s post. Or indeed of MIT students with some philosophy training and MIT students without such training.
Yes, it says that you could do better with either specific rationality training, or engineering training.
I don’t think you can take MIT student scores on the CRT as an indication of the effect of engineering training on rationality, any more than you could take NYU philosophy grad (NYU has the most prestigious philosophy program) scores as an indication of the effect of philosophy training.
This is true, but I’m still not sure how you get from this that engineering training is better for rationality improvement than philosophy training, given that Harvard and Princeton students are already well above the baseline score for average undergraduates.
Assuming there is no selection effect (and this is a pretty big assumption), philosophy training raises the CRT score of the average undergraduate from 0.65 to 1.16. Assuming engineering training accounts for the entire CRT score difference between Princeton and MIT students (another big assumption), engineering training raised their average score from 1.63 to 2.18. How am I supposed to draw conclusions from this data about which training is better for rationality?
I think (a) that there is a probably a big selection effect, and (b) that it is possible that the test used is biased in favor of mathematical rather than generally logical thinking. The CRT is also doesn’t include things like noticing when a word is meaningless, which I would think would be one of the most important skills for philosophers. I’m not sure how one would test that.
I think you’re right that the data don’t show what I had thought. I had thought that professional philosophers did worse than than MIT undergrads, but now it looks like there isn’t data about that. I think I was confusing it with the results from professional American judges (almost all graduates of the other program which claims to teach reasoning).
Sure, your average philosophy grad student is not as rational as your average LW meetup attendant or your average MIT student, but the evidence does suggest that they are more rational than your average grad student (assuming the CRT is a reasonably reliable test of rationality, of course). Surely that says something about the benefits of philosophy training. There are pretty strong selection effects associated with attending LW meetups and going to MIT. If that’s your bar for passing the CRT, it’s a very high bar.
I’d also note that looking at all people with some graduate training in philosophy is casting a pretty wide net. There are a lot of pretty bad philosophy graduate programs out there. I’d be interested in seeing a comparison of, say, students at the top 20 philosophy graduate programs with the groups mentioned in Luke’s post. Or indeed of MIT students with some philosophy training and MIT students without such training.
Yes, it says that you could do better with either specific rationality training, or engineering training. This is Luke’s point.
Me too!
I don’t think you can take MIT student scores on the CRT as an indication of the effect of engineering training on rationality, any more than you could take NYU philosophy grad (NYU has the most prestigious philosophy program) scores as an indication of the effect of philosophy training.
Compare MIT’s scores to Harvard’s or Princeton’s—they’ve all got super-smart students, but MIT does much better.
This is true, but I’m still not sure how you get from this that engineering training is better for rationality improvement than philosophy training, given that Harvard and Princeton students are already well above the baseline score for average undergraduates.
Assuming there is no selection effect (and this is a pretty big assumption), philosophy training raises the CRT score of the average undergraduate from 0.65 to 1.16. Assuming engineering training accounts for the entire CRT score difference between Princeton and MIT students (another big assumption), engineering training raised their average score from 1.63 to 2.18. How am I supposed to draw conclusions from this data about which training is better for rationality?
I think (a) that there is a probably a big selection effect, and (b) that it is possible that the test used is biased in favor of mathematical rather than generally logical thinking. The CRT is also doesn’t include things like noticing when a word is meaningless, which I would think would be one of the most important skills for philosophers. I’m not sure how one would test that.
I think you’re right that the data don’t show what I had thought. I had thought that professional philosophers did worse than than MIT undergrads, but now it looks like there isn’t data about that. I think I was confusing it with the results from professional American judges (almost all graduates of the other program which claims to teach reasoning).
Really? Philosophicalgourmet has told me otherwise. I’m interested in seeing a link or source, since I’m looking at programs ATM.
NYU is no. 1 on the Gourmet Report rankings.
Perhaps it’s not no. 1 for your particular field.
Woah, brain glitch. Sorry.