On Thu, 2 Feb 2006 20:05:14 -0800 [EMAIL PROTECTED] (Alex Martelli) wrote: > Benji York <[EMAIL PROTECTED]> wrote: > > Alex Martelli wrote: > > > Disagree -- far more people THINK they're clever, than > > > really ARE clever. According to a recent article in > > > the Financial Times, over 40% of a typical financial > > > firm's employees firmly believe they are among the 5% > > > best employees of the firm -- and the situation, > > > believe me, is no different in programming. > > > > It's apparently no different anywhere: > > http://www.phule.net/mirrors/unskilled-and-unaware.html
I have to say, I've read this paper, and it's pretty bad science. I think they seriously over-interpreted their data. There are a lot of different ways to interpret the data they present, and it takes a fair amount of sophistry for them to come up with it "supporting" their hypothesis. Which doesn't *disprove* the idea, but I remain unconvinced by their reasoning. No doubt the effect they describe is not non-existant, but I question whether it has as much impact as proposed. Note also in the datasets that *overperformers* *underrated* their performance. Also, the research was done on *particularly* subjective subject matter! Take the "humor" category for example. I also think this is precisely the kind of experiment that has to be made *very* objective to be at all useful. It's so easy to sensationalize: it plays on your self-worth fears, your cynicism, and your arrogance *at the same time*. It's hard to imagine that anyone accepting the results is doing so for scientific reasons. Frankly this paper sounds like a bid for the "Journal of Irreproducible Results" that somehow got accidentally submitted to a serious journal (of course, I don't know enough about psychology journals to know if that is really a "serious journal", but I'm taking it at face value). Of course, you *could* conclude that I'm just "unskilled and unaware of it" for believing this -- but that's the beauty of all self-reinforcing delusions, isn't it. ;-) Cheers, Terry -- Terry Hancock ([EMAIL PROTECTED]) Anansi Spaceworks http://www.AnansiSpaceworks.com -- http://mail.python.org/mailman/listinfo/python-list
