Design and Evaluation of a Programming Tutor Based on an Instant Messaging Interface

Abstract

Computer programming skills are essential for a variety of disciplines in the fields of Science, Technology, Engineering and Mathematics (STEM). To support learning of programming in personalized ways, Programming Tutors (PTs) have been utilized in higher education contexts for decades. However, implementation of PTs with mobile devices has remained unexplored from both design and adoption standpoints. In this research, we designed, implemented, and trialled a PT based on Telegram Messenger. The tutor was introduced to a cohort of engineering freshmen (N= 227 ) in a computer programming course; however, it was introduced as an optional learning tool with no extrinsic incentives for fostering its adoption. Under these conditions, students’ activity with the tool was monitored for 11 days. A total of 99 students (44.7 % ) chose to use the tutor and did so at least once. In average, students who used the tutor solved 9.2 tasks out of a maximum of 18 (SD= 5.49 ). The use of notifications was found highly influential to motivate tutor use: within 3 days of a student receiving a reminder notification, 80.6% of responses to tasks were registered. The tutor scored a mean of 73.8 (SD= 13.81 ) in the Systems Usability Scale. Students suggestions for improvement emphasized the need for more elaborate feedback after submitting their responses to problems and leveling of task difficulty. Regarding the tutor’s best features, mobility and ease of use of the interface were found the most prominent.