Table of Links
2. Contexts, Methods, and Tasks
3.1. Quality and 3.2 Productivity
5. Discussion and Future Work
5.1. LLM, Your pAIr Programmer?
5.2. LLM, A Better pAIr Programmer?
5.3. LLM, Students’ pAIr Programmer?
6. Conclusion, Acknowledgments, and References
4.2 Compatibility
Salleh et al. [70] listed multiple factors for pair compatibility, such as personality, perceived skills, actual skills (expertise), self-esteem, gender, and work ethic. Thomas et al. [81] found that paired students with similar self-confidence levels produce their best work. Hannay et al. [30] found that Big Five personality traits only have modest predictive value on pair programming performance, and expertise, task complexity, and country have stronger prediction power in comparison. There also seems to be evidence that women benefit from pair programming more than men do [29, 67].
Expertise as a compatibility factor has been extensively studied in the human-human pair programming literature. For example, researchers found that a student pair performs the best when their expertise is similar [70] and students preferred to be paired with similarly skilled partners [16]. However, in industry, Jensen [36] reported that when both members were near the same capability level and strongly opinionated, the collaboration was counter-productive and troublesome.
In the introductory programming context, Lui and Chan [45] found that pairing up novices results in a larger improvement in productivity than pairing up experts. However, there are concerns about the risk of “the blind leading the blind” if they don’t have an expert to consult with [4]. Researchers also found that less-skilled students learn and enjoy more than more-skilled students in pair programming [16, 47]. However, when the knowledge gap is too large, students can be less satisfied and the benefits of quality may be smaller [60]. Chong and Hurlbutt [17] reported that a novice programmer collaborating with an expert may become disengaged, have lower self-esteem, and be afraid of slowing down or annoying their more-skilled partner [4].
Authors:
(1) Qianou Ma (Corresponding author), Carnegie Mellon University, Pittsburgh, USA ([email protected]);
(2) Tongshuang Wu, Carnegie Mellon University, Pittsburgh, USA ([email protected]);
(3) Kenneth Koedinger, Carnegie Mellon University, Pittsburgh, USA ([email protected]).
This paper is
