16 November 2004Evaluation of Academic Positions at the University of Iceland
Evaluating the scientific credentials of applicants for academic positions is always a tricky business at the best of times. Moreover, as with any other human activity, it is inherently biased and subjective. Given roughly equivalent resumes, choosing one candidate instead of another one is often not done using purely scientific criteria, and we are all aware of this.
Apparently, however, the University of Iceland aims at a purely objective evaluation of academic credentials, and they have devised a glorified form of bean counting that is used to associate a numerical value with a resume. One's worth as an academic, and one's wage, is thus captured by this "magic number"! (It'd be great if they had a web site that takes as input the LaTeX source of one's resume, and would produce as output one's "University of Iceland" number to list alongside our Erdös number. This would make a nice programming project for our students.) This reminded me of George David Birkhoff's attempt at developing a mathematical theory of aesthetics which he applied to art, music and poetry presented in the book Aesthetic Measure (1933).
It this an objective criterion? Of course not! The bias is, not so carefully, hidden in the formula used to compute the magic number. This formula was generated by academics who stood to benefit by getting a good number out of it. A couple of hearsay examples will suffice to describe this bias. From what I have learned, a paper published in a top notch international journal like Information and Computation counts only little more than one published in the Icelandic scientific journal. One gets nearly as many points for each talk one delivers as for being in the PC of an international conference or chairing one. Papers published in refereed conference/workshop proceedings or journals published electronically do not give points.
I won't bore you with further examples. My message is that our life's work cannot be quantified like that!
Last modified: Wednesday, 17-Nov-2004 14:04:07 CET.