Multi-variant assignment generation and assessment techniques

Abstract

Approaches to generation of individualized assignments, based on various diversification models, are considered. The techniques are proposed and discussed for constructing a sufficient number of assignment variants of the same complexity, aimed at increasing their diversity and decreasing the possibility of cheating. The built-in automatic assessment function clears the instructor of the tedious work of looking through a large number of similar assignment papers. On the other hand, automatic self-assessment increases the efficiency of students' home work, reduces the turnaround time, i.e. the duration of the cycle"assessmentcorrection" after the initial completion of the assignment paper by the student. © Springer Science+Business Media B.V. 2008.