Confidence

[Jorge Soberon Mainero]

The Mexican Comission on Biodiversity (CONABIO) has the same problem. We
request providers of information (i.e., museums) to categorize each
determination according to a simple code: determined by known world
expert; determined by local expert; determined by an ecologist, biologist
or other trained professional, but not taxonomy expert; and unable to
qualify the determination. This is simple and has been more or less
accepted, but obviously there are a number of problems. I believe that a
very interesting thing would be to get some empirical information of the
distribution of determination errors for different groups, museums, etc.
Does anybody knows anything about such hard data? I assume that the error
is binomial, but has someone fitted a curve?