Investigating the Effect of Student' Personality Traits Towards Improving Pair Programming's Effectiveness as a Pedagogical Tool for CS/SE Education

Abstract

Pair programming (PP) is a practice where two people sit side by side, using only one computer, and work collaboratively on the same design, algorithm, code or test. Given that each member of a pair presents their own personality traits, numerous studies have investigated the possible effect that these traits may have upon the pair's work. However, the results of a Systematic Literature Review carried out as part of this research showed that despite existing evidence of PP's effectiveness in a higher education context, previous PP studies presented inconsistent results in terms of the effect or influence of personality towards that effectiveness. In addition, the personality instrument that had been previously used was also criticized by personality psychologists for being unreliable in measuring an individual's personality traits. The aim of this research was to improve the implementation of PP as a pedagogical tool for use in Higher Education through understanding the impact that the variation in the personality composition of paired students has towards their academic performance. The personality measurement framework used in this research was the Five-Factor Model, which comprises five broad traits (Openness to experience, Conscientiousness, Extraversion, Agreeableness, and Neuroticism). We investigated the effects of personality composition on PP's effectiveness by conducting a series of formal experiments at the University of Auckland. We used as our subjects undergraduate students attending either an introductory programming course or an intermediate software design and construction course. This thesis reports the five formal experiments that separately investigated three of the five traits part of the Five-Factor Model, namely Conscientiousness, Neuroticism, and Openness to experience. These three traits were selected because evidence shows that they are educationally important and relevant for higher education. Our findings showed that two of the three personality traits investigated - Conscientiousness and Neuroticism, did not present a statistically significantly effect upon paired students' academic performance. However, our results revealed that Openness to experience played a significant role in differentiating paired students' academic performance. Our results also indicated that PP not only caused an increase in satisfaction and confidence levels but also brought enjoyment to the tutorial classes and enhanced students' motivation.