« Timeless music | Main | Protect them from themselves »

Interesting

26 Jan 2008 01:00 pm

Favorite books and average IQ. I find it most amusing that a web page titled "Books that make you dumb" has labeled Tuesdays with Morrie and The Five People You Meet in Heaven as works of philosophy.

Comments (16)

everything clumped around the middle is interesting?

and. Where does it mention IQ? It's SAT scores.

I find it amusing that there are books to the left of "I Don't Read."

Alex, SAT purports to measure something much like intelligence and highly correlates with measured IQ.

Megan, it also has Tuesdays with Morrie as fantasy or science fiction on the subset picture.

I sure doesn't have many books on the right hand end of the scale.

He's got Lolita as "erotica". Ooo-kay...

Charlie, Facebook doesn't have many users on the right-hand end of the scale. Could be worse, though. Could be MySpace.

I also noticed that he had Lolita as "erotica" and that he says that "English majors" had written in to complain of the categorisation.

The question that arises is that if this guy is so smart, why is he making up his own categorisations, rather than using the official Library of Congress categorisation (which I assume is determined by the publishers), or the one on Amazon (which I assume takes it data from that source)?

Another triumph of this guy's manual data handling is that he has "The Bible" and "The Holy Bible" as separate publications.

And the colors are all wrong! God, what was he thinking?!

Another triumph of this guy's manual data handling is that he has "The Bible" and "The Holy Bible" as separate publications.
Actually, if you read the site he has a very reasonable explanation for that.

Before everyone gets all uptight about genre categories and such, remember this is just a CS grad student doing some hacking, not serious social science research. There are flaws if you want to go looking for them, but a lot of them are due to things like Facebook users listing an author as a book. It's a cool toy. Let's try not to loose perspective.

By the way, this is the same guy who wrote WikiScanner.

I find it most amusing that a web page titled "Books that make you dumb" has labeled Tuesdays with Morrie and The Five People You Meet in Heaven as works of philosophy.

That made me laugh out loud.

Was there any discussion of whether listing a favorite book on facebook makes you smart or dumb, or how representative the practice was for college students?

Also, "Crime and Punishment" and "100 Years of Solitude" aren't anybody's favorite books - they're books you say are your favorite because you think it makes you look smart. I have more than a little difficulty buying the idea that those are favorites at CalTech, supposedly the smartest school.

Based on the "African American" cluster, I'm guessing Virg is a big Charles Murray fan...

Actually, if you read the site he has a very reasonable explanation for that.
Actually, he has a terrible explanation. He's already manually categorising books into genres. As a "CS Grad" he has no excuse for not using freely available metadata.

As to the difference between the data, all we see is that there are fewer entries for "the holy bible" than "the bible."

Charlie, my IQ and SAT score deviated quite widely. I'd once again be curious if that was just a typo, or if Megan intends to conflate IQ with SAT score, which would be bunk.

Charlie, my IQ and SAT score deviated quite widely. I'd once again be curious if that was just a typo, or if Megan intends to conflate IQ with SAT score, which would be bunk.

Considering my SAT score was about 300 points higher than my husband's, and we have nearly identical IQs (according to online tests), I would concur that SAT and IQ don't correlate as much as some people think.

I think it's perfectly obvious that people who list "The Bible" are going to have different social characteristics than those who list "The Holy Bible". Considering that there's a 150-point SAT difference between the two, and the error bars are small, I think he's found one of those differences.

His classification by genre doesn't actually show that much, so errors in his classification don't mean much.