ITSAS: An approach towards adaptive student assessment

Traditional assessment model uses number of correct responses given by a student as a sole parameter to evaluate the performance of a student. In this paper we have suggested a new scheme to determine the difficulty of questions and an improved evaluation criterion which uses hardness of questions, several student's behavioral factors and attributes of a quiz. We have also proposed an adaptive assessment model which used student's performance and time spent in a quiz to predict his/her expertise level. This assessment model is implemented in a prototype which gives personalized quizzes to the students based on their expertise level. Empirical results have shown that the student's real performance in paper-based quizzes matched with their predicted performance level and thus it proves our hypothesis.