Abstract

ROQUE HERNANDEZ, Ramón Ventura et al. Comparison of Pair and Solo Programming through software metrics in University Students’ Projects. RIDE. Rev. Iberoam. Investig. Desarro. Educ [online]. 2019, vol.10, n.19, e030.  Epub Nov 17, 2020. ISSN 2007-7467.  https://doi.org/10.23913/ride.v10i19.549.

Introduction: Pair Programming is an agile practice that can be used both in software development for business and in the teaching of programming in university courses. Objective: This paper presents a research that was conducted to compare pair programming and solo programming in university courses from the perspective of the metrics of the programs created by freshmen enrolled in the Bachelor Degree in Information Technologies. Method: The participants were divided into two groups: those who applied pair programming and those who programmed individually. Both developed the same program under the same conditions. The following metrics were analyzed in their programs: Number of Statements, Percentage of Comments, Maximum Depth, Average Depth, Maximum Complexity, Number of methods per class, Number of calls per method and Number of sentences per method. The values of the metrics were obtained by the Source Monitor software. Then, Mann-Whitney tests were performed in SPSS. Results: Results showed that students that worked in pairs wrote code with more statements (p=0.038, U=17.00) and a higher level of depth (p=0.032, U=18.00) compared to solo programmers. Conclusions: This paper contributes to the field of software development teaching by providing quantitative empirical evidence on the effectiveness of pair programming. It is concluded that pair programming can be an appropriate educational approach for the beginner’s software development university courses.

Keywords : Computer Programming; Software; Higher Education; Measurement.

        · abstract in Spanish | Portuguese     · text in English     · English ( pdf )