Acceptance and Assessment in Student Pair-Programming: A Case Study

Abstract

The learning of programming is a complex process that may lead to high failure and dropout rates. Pair Programming is a software development technique that involves two individuals working side-by-side using the same computer. It has proven to be helpful in the process of programming learning. This paper analyses pair programming's acceptance and assessment in novice programming university courses considering participants' gender, previous programming experience and programming enjoyment. The participants were 80 students from three different cohorts enrolled in the IT undergraduate program at a Mexican State University. We designed and administered a questionnaire to collect data after the pair programming exercises were performed. For data analysis, we used SPSS 24, Mann-Whitney, Kruskal-Wallis and Jonckheere-Terpstra statistical techniques. Descriptive and comparative results showed a significant increasing monotonic trend in the acceptance of Pair Programming as students' preference towards programming increased (standardized statistic = 3.20, p = 0.00, Kendall's b = 0.30, p = 0.001). There were no other statistically significant results. We found Pair Programming helpful and easy-to-implement in university courses. Students, on the other hand, positively accepted and assessed this learning approach. We recommend Pair Programming adoption in university courses as part of a broader strategy to engage students in the effective learning of programming.