If there is more than one evaluation criterion, then finding the x% best of a finite set of alternatives is a Multi-Criteria Decision Making (MCDM) problem where the decision making process  usually suffers from the presence of non-comparable alternatives: neither is A better than B, nor B better than A w.r.t. all criteria.

The easy way-out in such situations is the use of a real-valued function to rank the alternatives:  a rank function, a utility function, a scalarization method such as weighted sums, an indicator - it goes by many names. Among other things, the Dean's list at our faculty is compiled in such a way.

In the three parts of the seminar, a new ranking method coming from multivariate statistics is presented (Kostner), the Dean's list rankings according to the old and the new method are compared (Steinbach, based on her Bachelor thesis) and some ramifications discussed (Hamel).

A major finding is that our current Dean's list puts students at a disadvantage who try to achieve the 60 CPs per year required to complete a Bachelor program in favor of students who focus on fewer subjects/CPs while the new ranking method does not have this effect.

For online partecipation, please register in advance at the link below.