A genetic algorithm‐based approach for making pairs and assigning exercises in a programming course

Abstract

Pair programming is an approach where two programmers work to solve one programming problem sitting shoulder to shoulder on a computer. Several studies indicating numerous benefits of using pair programming as a teaching strategy exist. However, only a few of them take into consideration the mechanism followed for pair formation. With an aim to study the impact of pair programming on undergraduate students, we try to make the pairs compatible with a genetic algorithm-based approach. Using a genetic algorithm, the system ensures that every pair in the class gets a particular combination of skills and personality traits. We also developed a desktop application to assign programming exercises to students dynamically. To assess the efficacy of pair programming in introductory programming course, a formal pair programming experiment was run at Netaji Subhas University of Technology. The pair programming experiment involved a total 171 undergraduate students from a computer engineering course. At the end of the program, we assessed the programming abilities of every student. We also analyzed the impact of a genetic algorithm-based pairing mechanism. On the basis of assessments, it is observed that pair programming is a successful pedagogical tool for facilitating active learning of introductory programming courses. Responses to survey garnered from undergraduate students hint that the genetic algorithm approach leads to compatible pairs.

Fostering computer programming and digital skills development: An experimental approach

“…Lastly, this study also supports the findings of Jung et al [75] and Tehlan et al [147] who found that students improved in their study outcomes as a result of the change in instructional design. Perhaps, the use of innovative pedagogy like CBL and PBL is responsible for the treatment group's active participation leading to improved performance in a computer programming course.…”

The application of AI technologies in STEM education: a systematic review from 2011 to 2021

“…Mathematics (N = 3) attracted the least attention. In addition, 7 studies contained interdisciplinary subjects, such as computer engineering (Buenaño-Fernández et al, 2019;Tehlan et al, 2020), engineering mathematics (Deo et al, 2020), and integrated STEM education (Suh et al, 2019;Wang, 2016).…”

The effects of online peer-facilitated learning and distributed pair programming on students’ learning

“…Pair programming is an approach of collaborative learning in the field of computer science which involves two students working together on a programming project (Bowman et al, 2020). The collaborative nature of pair programming could lead students to achieve advanced learning performance in terms of efficiency and quality ( Kanika et al, 2020 ; Sobral, 2020 ), to be more confident, and to gain better grades on their programming assignments ( Xinogalos et al, 2019 ); moreover, it is regarded as a successful teaching method and tool for facilitating students’ active learning in programming courses ( Sun et al, 2021 ; Tehlan et al, 2020 ).…”

“… Dominic et al, 2020 ; Xinogalos et al, 2019 ), few of these studies investigated its effects on improving non-computing majors' learning outcomes. In addition, only a few studies take into consideration the mechanism of pair programming or distributed pair programming followed for pair formation ( Tehlan et al, 2020 ). As distributed pair programming is recently drawing attention among educators and researchers ( Demir & Seferoglu, 2021 ), its effects should be investigated.…”