The Effects of "Pair-Pressure" and "Pair-Learning" on Software Engineering Education

Author(s): Williams, L., Kessler, R.R.
Venue: Conference of Software Engineering Education and Training
Date: 2000

Type of Experiement: Case Study
Sample Size: 20
Class/Experience Level: Undergraduate Student
Data Collection Method: Survey, Project Artifact(s)



How Pair Programming Was Used: Students learned and programmed in ASP. They were paired at the beginning of the semester and did all classwork together.

This is a longer, more complete paper on the same study done in "But, isn't that cheating?" by Williams. It begins with an introduction to Pair Programming, and benefits seen in industry. Then it introduces the course, and some results from it. They found that the collaboration kept students on track (ie: they were less likely to surf the web or read emails), and that it "served as an extremely efficient defect removal method". Students felt the pressure of working with a partner increased their productivity and the quality of their work. Additionally, the teacher benefitted: students happier, assignments on time and high quality, fewer questions to the teacher (partners were often able to answer questions), and fewer partner problems than in a divide and conquer partnership.