Empirical Confirmation of Pair Programming Effectiveness in the Teaching of Computer Programming

Author(s): Bassey Isong, Thuso Moemi, Nosipho Dladlu, Neo Motlhabane, Ohaeri Ifeoma and Naison Gasela
Venue: 2016 International Conference on Computational Science and Computational Intelligence
Date: 2016

Type of Experiement: Case Study
Sample Size: 42
Class/Experience Level: Undergraduate Student
Participant Selection: All student in class
Data Collection Method: Observation, Code Metric, Project Artifact(s)


This paper investigates the effectiveness of pair programming in teaching introductory level computer programming courses at a Sophomore level in college. The study looked at 42 second year students in an Object Oriented Java class that was offered as an elective. During the experiment the authors looked at time to complete a task, how correct the code was, code quality, as well as student motivation in determining if pair programming is an effective means to teach computer programming to beginners.

While conducting the study there were two forms of data collection. The first was in the form of the teacher collecting and verifying the code as well as the time taken to complete the task. The second was in the form of an interview that was completed once the students were finished working on their project. The majority of students stated in the interview that they enjoyed working in pairs and that they learned from their partners during the process. The results showed that overall paired programming was an effective was to teach programming to beginners.