Dunning wondered whether it was possible to measure one’s self-assessed level of competence against something a little more objective — say, actual competence.
A much more interesting question, one for my next blog entry, is the opposite: in a complex world of millions of different specialties about which much is known at great depth, might it be impossible for anybody to know when one is exceeding one's competence? Might this, instead of being a humorous story about bank-robbers or a story about "others", be a story about the entire population of the planet?
The easy question is: given a specific field, how do we help people self-rate? But it's not a very practical question, because in the real world it's never just one vertical field to rate inside of. The tougher question is the matrixed-skills one. The brain surgeon who speaks about car maintenance as an expert although he is completely wrong, the psychologist that waxes on about social ills as an expert although he is completely off-base, the college professor in mathematics that forays into economics with nothing more than a lunch pail and a bologna sandwich. This is the really the more interesting (and important) scenario. (The hacker that ventures into cognitive psychology. Yes, I get the irony.)
Note that I picked skills that were far apart from each other: brain surgery and automotive repair, for instance. These are the easy cases. The crazy hard ones are where the skillsets are very close to each other. Not sure that outsiders could spot that happening.
There are actually two different questions about competence:
Competence compared to some objective standard, basically "Can I do that?", which I think often can be answered realistically (at least where "that" is specific enough).
Competence compared to others, or how do my abilities rank compared to others, which may not really be answerable, except very generally and vaguely.
The big problem comes in trying to generalize from specific competencies to "general competence".
A much more interesting question, one for my next blog entry, is the opposite: in a complex world of millions of different specialties about which much is known at great depth, might it be impossible for anybody to know when one is exceeding one's competence? Might this, instead of being a humorous story about bank-robbers or a story about "others", be a story about the entire population of the planet?
The easy question is: given a specific field, how do we help people self-rate? But it's not a very practical question, because in the real world it's never just one vertical field to rate inside of. The tougher question is the matrixed-skills one. The brain surgeon who speaks about car maintenance as an expert although he is completely wrong, the psychologist that waxes on about social ills as an expert although he is completely off-base, the college professor in mathematics that forays into economics with nothing more than a lunch pail and a bologna sandwich. This is the really the more interesting (and important) scenario. (The hacker that ventures into cognitive psychology. Yes, I get the irony.)
Note that I picked skills that were far apart from each other: brain surgery and automotive repair, for instance. These are the easy cases. The crazy hard ones are where the skillsets are very close to each other. Not sure that outsiders could spot that happening.