Fair exams in split groups: (Implementation of equitable computer based exams for large groups in small test centers)

How to provide a fair computer based exam for a large group, if only a small number of computers is available? Just sequencing sub groups in exam time slots does eliminate the option to use the exact same test for all groups. Using different tasks for all groups may result in different difficulty levels for the consecutive groups. This paper describes the principle of a rating system as part of a web based exercising platform, in which the rating outcomes can be used to provide different, but equally difficult tasks for sequenced exam groups. An important factor in this is the combination of subjective, personal feedback and clinical, algorithm based data in the rating system.