Abstract

The availability of unobtrusive recording of human-computer interactions is one of the many advantages of computerised testing. The paper reports on results of the research in test takers behaviours in the course of implementation of online formative and summative assessment. The study examines statistics of self-assessed confidence, and its correlation with accuracy data collected from computerised tests in basic mathematics skills administered to tertiary students in 2005-2006. Practical recommendations on how self-assessment of confidence can be useful to learners and teachers will be discussed. Test designers, education practitioners and cognitive researchers may benefit from the results of the study.