Assessment is an important component in teaching where a student is assessed for his knowledge and skills in the concepts that were taught to him. Assessment strategies include written examination, viva voce, seminar, assignments, projects, etc., and are often based on assessor’s intuitive understanding of what has to be assessed, and thus remains largely subjective in nature.
Written examinations comprise questions that are of different types: objective, multiple choice, True/False, essay, problem solving and design. Marking techniques in written examinations primarily focus on the correctness of results often using numerical values. A score of 0 is offered for wrong answers, and 1 (full mark) for correct answers. (We refer to this as 0/1 marking.) Partial marking is more exhaustive and fairer than 0/1 marking since an incorrect answer can be partially correct. However, partial marking strategies are not based on formal techniques, and may not be uniformly consistent. Techniques to improve objectivity in marking include multiple choice, short answers, and modularizing large questions into multiple smaller questions minimizing their dependency on each other. However, the format and structure of the questions may not permit effective partial marking. Existing automated tools at best provide passive support to assessment and they do not explicitly model the concepts students learn in their curriculum.
There is thus a need for providing a methodology for a more accurate method of assessing a student’s answer in written examinations.
