On Developing and Communicating User Models for Distance Learning Based on Assignment and Exam Data
Students who enrol in the undergraduate program on informatics at the Hellenic Open University (HOU) demonstrate significant difficulties in advancing beyond the introductory courses. We use decision trees and genetic algorithms to analyze their academic
- PDF / 252,019 Bytes
- 19 Pages / 439.18 x 665.994 pts Page_size
- 49 Downloads / 174 Views
Hellenic Open University, Patras, Greece, {thh, kalles, pierrakeas}@eap.gr Research Academic Computer Technology Institute, Patras, Greece
Summary. Students who enrol in the undergraduate program on informatics at the Hellenic Open University (HOU) demonstrate significant difficulties in advancing beyond the introductory courses. We use decision trees and genetic algorithms to analyze their academic performance throughout an academic year. Based on the accuracy of the generated rules, we analyze the educational impact of specific tutoring practices. We examine the applicability of these techniques in a senior course and reflect on some software engineering issues involved in the development of organization-wide measurement systems. All results are based on data drawn from academic records.
8.1 Introduction All measurements affect that which is being measured. However, measurement for performance evaluation greatly affects what is being measured and may distort the process being evaluated. Such distortions make the education field particularly vulnerable due to misleading policies. For example the number of computers per 100 high school students is an indicator of educational ICT utilization (other things being equal). When however this indicator is used as a funding goal (“by 2007 all EC member states should have 1 computer per 10 students”) it results in schools keeping old machines and in the total ICT budget being shifted toward hardware. It is reasonable to suggest that student success is a natural success indicator of a University (of a teacher, of a class, or of a course). However, if that success is used as a criterion for tutor contract renewal, and if students must evaluate their own teachers, then tutors may tend to lax their standards. This chapter is about dealing with this issue in the context of a large (over 25,000 students) Open-and-Distance-Learning (ODL) University, the Hellenic Open University (HOU) and in particular its undergraduate Informatics program. We ask how we can detect best distance tutoring practices and associate them T. Hadzilacos et al.: On Developing and Communicating User Models for Distance Learning Based on Assignment and Exam Data, Studies in Computational Intelligence (SCI) 104, 137– 155 (2008) c Springer-Verlag Berlin Heidelberg 2008 www.springerlink.com
138
T. Hadzilacos et al.
with measures of students’ success in an “objective” way and, subsequently, effectively disseminate this information to all interested parties. The measurement strategy we have developed to-date in HOU has been progressively refined to deal with two closely linked problems: that of predicting student success in the final exams and that of analyzing whether some specific tutoring practices have any effect on the performance of students. Each problem gives rise to the emergence of a different type of user model. A student model allows us, in principle, to explain and maybe predict why some students fail in the exams while others succeed. A tutor model allows us to infer the extent to which a group of tutors
Data Loading...
