Decision trees in formative procedural knowledge assessment

Abstract

In this paper, an approach to automated formative assessment of procedural knowledge is described and evaluated. While assessment and representation of conceptual knowledge using visual aids such as of concept maps has often been analyzed and discussed in the literature, significantly less attention has been given to assessment of procedural knowledge. The approach described in this paper is based on automated evaluation of knowledge described by examinees in form of a decision tree against a repository of test cases. The examinee is afterwards provided with automatically generated feedback about the overall results of applying his decision tree, as well as possible changes required to correct them. The results of the pilot evaluation of a prototype system implementation are described and discussed.