Colbert May Not See Race . . .

Ian Ayres

but Jane Fonda and I can tell you that he definitely sees sex, the sex of his guests that is.

In an earlier post, I reported on some number crunching I did concerning Colbert interview questions. I reported on differences in the kinds of questions that Colbert asks people who are identifiably liberal or who are presented as experts. And I challenged you to try to predict whether he asked women different questions than men, or "famous" guests different questions than non-famous guests.

Once again there is evdience of "discrimination."

Colbert was 22.3% less likely to ask women guests negative questions (p. < .05) and 40.1% less likely to ask women guests questions that take a premise to a logical extreme (p. < .01).

Okay Colbert nation, even now you can play the prediction game. When the guest is "famous," do you think he is more or less likely to frame a question as a statement? [If it helps, I reported last time that Colbert was 15% less likely to frame questions as statements when the guest was an expert.] So what do you think?

PMS Chicago guessed "he'll take a premise to a logical extreme more often with famous guests than with non-famous ones." You are correct! Colbert was 11.7% more likely to ask "extreme" questions to "famous" guests (but this result is only statistically significant at p. < .1).

Turns out in my (very limited) sample, that the 11th child was also 12.2% more likely (p. < .08) to frame questions as statements when the guest was identifiably famous.

As I said before, this is NOT serious Super Crunching. The coding was fairly arbitrary. (The process of coding was enjoyable and led to some pleasant discussions with my co-coder and son Henry). I did it because I've been doing a lot of interviews for the book and interviewers generally have a hard time thinking that number crunching could ever be applied to what they do. The point isn't that we should trust these particular results. The point is to help us imagine what a more serious attempt might be able to do.

So here's a final challenge. Can you think of a way that number crunching could help you do your own job better? I predict that your first answer will be "no." But part of the point of my book is to suggest that we all should ask ourselves that question again.
Posted 12:28 PM by Ian Ayres [link]