Information technology skills assessment techniques

Abstract

There are two tasks associated with the assessment of IT skills; the discovery of errors and the classification of errors. The discovery of errors is a pattern matching task where a candidate attempt is synchronised with the model answer. The classification of the errors consists of applying sets of rules to regions of a candidate answer corresponding to regions of the model solution. A particular constraint in this type of assessment is that the results should be similar to those which would be produced by a human examiner. This paper describes the algorithms the authors have used in automating a number of assessors for the use of word processors, spreadsheets and databases and describes some of the problems of applying such algorithms in practice.