THE INFLUENCE OF DYNAMIC HOMEWORK ASSIGNMENTS ON STUDENTS' LEARNING BEHAVIOR

At the Vienna University of Economics and Business (WU) there are about 5.000 freshmen per year, who have to attend a standardized one-year introductory phase, after which they choose their majors. The introductory phase is tightly structured: in each half semester, during a five-week period students attend classes (twice a week for each course). The sixth week without classes should be used for intensive learning and repeating, and is followed by the final exam.

The courses in this introductory phase are taught in large groups of about 170 students and are supported by the university’s e-learning platform. The courses’ final exams are organized as multiple choice exams which are evaluated automatically.
Due to the organization of the introductory phase, the tight time schedule, the number of courses which should be completed and the organization of the exams, students tend to learn at short notice one or two weeks before the exam, and therefore tend to forget the relevant topics soon afterwards.

To give students incentives to keep up with the classes they attend, we introduced voluntary dynamic homework assignments in the course “Introductory Finance” last year. Students can solve those exercises via the e-learning platform and gain some additional points for their final exam. (The term “dynamic” means that the numbers and even some text modules and multiple choice answers are modified within certain ranges for each user.)

We analyze students’ learning behavior using the logfiles of the e-learning platform and find that the dynamic homework assignments serve their purpose. Preliminary results show that, averaged over several exams, with homework assignments students start to learn earlier (three weeks before the exam instead of one week), learn more often (8 days instead of 5 days) and solve more exercises at the e-learning platform (223 instead of 147).

Examining the data before and of one single exam, it also turns out that there is a highly significant statistical difference between the groups of students who solve at least one homework assignment and those who don’t. The one-factor analysis of variance state that the first group achieves approximately 10 percent more points at the final exam.
Therefore (dynamic) homework assignments are an useful tool to counteract short-term learning behavior tendencies. Based on an e-learning platform, they can be easily employed for mass lectures, where those learning strategies are more inherent than in small seminars.