Auto-marking System: A Support Tool for Learning of Programming

Computer programming requires skills in designing algorithms, understanding syntax, writing programs, as well as the ability to correct errors in order to produce good programs. These skills can be developed through a lot of practice on a continuous basis. The students’ proficiency in programming is measured by the number of exercises that can be solved correctly within a specified time frame. From past observations, it is discovered that most students were able to solve the problems given during laboratory sessions. However, their performances did not carry over to laboratory tests. This situation points to the possibility that the students might have not perform adequate self-practice in preparing for laboratory tests. In a student-centred learning environment, fulfilling the notional learning hours is essential to ensure that students are prepared to take their subsequent classes. Based on a constructivist learning framework, this article reports the development and evaluation of a prototype system to assist in self learning of programming. The online Auto-marking Programming Exercise System was developed based on the UVa Online Judge as benchmark. The system can provide real time feedback to students immediately after the students submit their programs. This instant feedback is an important characteristic of the constructivist approach to learning. This will help students learn programming in an effective way. The system is tested and evaluated for usability by selected users from among instructors and former students of computer programming course