I think you're completely mis-characterizing both what I said and why I said it. My complaint isn't that I don't like interviews, or think they're uncomfortable, or think quantitative metrics are better (If I were a school I would definitely pick someone who was interesting with lower stats over someone boring with high stats), or even that there are better ways than an interview to pick an applicant (I'm not sure there are). My point was merely that an interviewer should not take the interview as a definite example of a person's character (like you said, writing down pompous, arrogant, etc. These things simply cannot be determined in an hour).
On the topic of business interviews, again that wasn't at all what I was getting at. I'm not saying that businesses don't place emphasis on interview, I'm saying precisely the opposite--that they DO place an ENORMOUS emphasis on interview, and that they take the interview to be be indicative of job performance (which in something like advertising or even a managerial position, is highly dependent upon the skills that people think are being tested in an interview, e.g. how well people get along with others). My point is that the interview simply isn't indicative of job performance in a particularly appreciative extent and that this emphasizes that the interview does not do an in any way reasonable job of determining character traits in people. A correlation of .1 is barely more than what would be expected from random chance, and so you can say that all an interview is able to do is judge people who are complete misanthropes or psychopaths or something. But among the vast majority of people, an interview cannot differentiate them.
When i say the methodology used is almost meaningless does not mean I have in mind any conception of an adequate methodology. In fact, I'm quite skeptical about whether one is reasonably possible. I realize people can't tell how well they did on interviews, and so on and so forth. I realize that everywhere uses a similar process as medical schools. But I disagree when you say it works, because the simple fact is all the evidence available (and there have been numerous studies on this and other things related) shows that it does not work, and that everyone thinks they do. Again, I'd like to emphasize that I have nothing to be bitter about in terms of rejections post-interview, I enjoy interviews, etc. But when people say, "oh, you can tell this or that about a person, if you do an interview, you'll know," they are deluding themselves about our ability to judge another person's character. That's why they say "first impressions count," etc. Because people jump to conclusions, taking no evidence at all (something like a person's hairstyle) to say something about their character. Interviewers need to be vigilant of these problems.
