Investigating Strategies used by Novice and Expert Users to Solve Parsons Problems in a Mobile Python Tutor

Abstract

We present PyKinetic, a mobile tutor for Python. The tutor is aimed at novices and meant to be a complement to traditional lectures and labs. The first type of activities implemented in the tutor is Parsons problems, which present code snippets to be ordered by the student to produce the desired output. As a starting point towards an intelligent tutor, we conducted a pilot study to evaluate the interface and usability of PyKinetic, and to identify and contrast strategies used by novice learners with those of experts. Great feedback and enthusiasm was received for the prospect of PyKinetic and interesting strategies were revealed from both groups. The study revealed that experts, as can be expected, outperformed novice users and used superior problem-solving strategies, In future work, we will improve PyKinetic's problems, feedback, activities and extend PyKinetic to provide instruction on optimal problem-solving strategies.