This paper applies Selten's measure of predictive success to the probl
em of comparing and ranking various theories of decision making under
risk. Two experiments were carried out for this application: one a Pai
rwise Choice experiment involving 15 such choices and the other a Comp
lete Ranking experiment involving 11 risky choices. The results show t
hat there are difficulties of both a practical and a philosophical nat
ure in applying Selten's measure. (C) 1998 Elsevier Science B.V.