A method is proposed for predicting the pass rate of a Computer Science course. Input data comprises different software metrics that are evaluated on a set of programs, comprising students’ answers to a list of computing challenges proposed by the course instructor. Different kinds of uncertainty are accepted, including missing answers and multiple responses to the same challenge. The most informative metrics are selected according to an extension to vague data of the observed Fisher information. The proposed method was tested on experimental data collected during two years at Oviedo University. Yearly changes in the pass rate of two groups were accurately predicted on the basis of 7 software metrics. 73 volunteer students and 1500 source files were involved in the experimentation.