Authors:
Lisa van der Heyden
1
;
Fatma Batur
2
and
Irene-Angelica Chounta
1
Affiliations:
1
Department of Human-centered Computing and Cognitive Science, University of Duisburg-Essen, Duisburg, Germany
;
2
Computing Education Research Group, University of Duisburg-Essen, Essen, Germany
Keyword(s):
Programming, Novices, Adaptation, Intelligent Tutors, Python, Errors, Misconceptions.
Abstract:
Students often struggle with basic programming tasks after their first programming course. Adaptive tutoring systems can support students’ practice by generating tasks, providing feedback, and evaluating students’ progress in real-time. Here, we describe the first step for building such a system focusing on designing tasks that address common errors and misconceptions. To that end, we compiled a collection of Python tasks for novices. In particular, a) we identified errors occurring during introductory programming and mapped them to learning tasks; b) we conducted a survey to validate our mapping; c) we conducted semi-structured interviews with instructors to understand potential reasons for such errors and best practices for addressing them. Synthesizing our findings, we discuss the creation of a tasks’ corpus to serve as a basis for adaptive tutors. This work contributes to the standardization and systematization of computing education and provides insights regarding the design of
learning tasks tailored to addressing errors.
(More)