Distributed Pair Programming Research Articles

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

The COVID-19 pandemic has forced universities and schools around the world to adopt online learning. Teachers may wonder if their students can attain satisfactory learning performance in an online learning environment without teachers' on-the-spot attention. In order to develop students' skills in programming, promote their enjoyment of learning and intention to learn to program, the researchers integrated two innovative teaching approaches, using online peer-facilitated learning and distributed pair programming, and investigated the effects of these on students' online learning performance. This study conducted an experiment that included 128 undergraduates from four class sections of Department of Finance. Thus, the experimental design in this research was a 2 (Peer-facilitated learning vs. non-peer-facilitated learning) × 2 (Distributed pair programming vs. non-distributed pair programming) factorial pretest/post-test design. The participants in this research mainly consisted of four classes of students from a non-computer or information department who took a compulsory course on programming design.’ Both quantitative and qualitative data were collected in this study. According to the results, the peer-facilitated learning group exhibited significantly better development of programming skills, enjoyment of learning, and intention to learn, than the non-peer-facilitated learning group. However, expected effects of enhancing the learning of the students in this study who received the distributed pair programming were not found. The design of online pedagogy can be a reference for online educators. The implications of applying online peer-facilitated learning and distributed pair programming to support students’ learning and the design of online programming courses are discussed.

Adopting distributed pair programming as an effective team learning activity: a systematic review.

As online learning has become an inevitable trend in the post-peak era of the COVID-19 pandemic, distributed pair programming (DPP) is gaining momentum in both education and industry. DDP serves as a collaborative programming approach and also benefits the development of computational thinking, a fundamental skill in today's world. This study conducted a systematic review of studies on DPP published after 2010 to understand the themes and factors that impact the team effectiveness of DPP and thus inform future research and practices on how to better leverage this approach for teaching and learning. The results showed that individual characteristics attracted major investigations in the selected 23 studies, including prior programming experience, actual skill, perceived skill, gender, personality, time management, confidence, and self-esteem, with pair compatibility identified as a critical team design factor that significantly affects programmers' satisfaction. Although the feel-good factor in the team process was investigated, no significant impact was found. Under the team environment theme, we compared different opinions on the orientation (e.g., scripted roles) and the use of technology (e.g., integrated development environment tools). Future research should investigate how task structure influences team effectiveness of DPP and relates to computational thinking education. Additionally, because most studies were conducted in higher education contexts, more research in primary and secondary educational contexts is also needed.

Distributed Pair Programming in Higher Education: A Systematic Literature Review

This paper presents a Systematic Literature Review (SLR) of fifty-seven studies on Distributed Pair Programming (DPP) in higher education, identifying which studies investigated factors on the effectiveness of DPP as a method for learning programming, factors related to mediating and stimulating interactions between students, the measures/instruments used for exploring these factors, as well as the tools and their features. As DPP effectiveness is very promising as regards code quality and academic performance, the findings can contribute to a better understanding of how communication, collaboration and coordination are investigated. It was found that there are very few studies concerning compatibility, pair formation, and role contribution, the majority of which did not use data derived from students’ actions. Only a small number of studies used DPP tools with logging capabilities. Even though several IDEs and Eclipse plugins have been designed for DPP and offer specialized features, there are still studies which use a combination that include video conferencing and remote sharing tools along with various IDEs and auxiliary tools. This SLR can be of interest to educators that aim to apply DPP in educational settings, researchers designing informed empirical studies, as well as designers of DPP tools.

An Empirical Study on Factors related to Distributed Pair Programming

This paper reports students’ perceptions and experiences attending an object-oriented programming course in which they developed software using the Distributed Pair Programming (DPP) technique. Pair programming (PP) is typically performed on one computer, involving two programmers working collaboratively on the same code or algorithm. DPP on the other hand is performed remotely allowing programmers to collaborate from separate locations. PP started in the software industry as a powerful way to train programmers and to improve software quality. Research has shown that PP (and DPP) is also a successful approach to teach programming in academic programming courses. The main focus of PP and DPP research was PP’s effectiveness with respect to student performance and code quality, the investigation of best team formation strategies and studies of students’ attitudes. There are still limited studies concerning relationships between performance, attitudes and other critical factors. We have selected some of the most common factors which can be found in the literature: academic performance, programming experience, student confidence, feelgood factor, partner compatibility and implementation time. The main goal of this study was to investigate correlations between these attributes, while DPP was used as the main programming technique.

Examining Student Performance and Attitudes on Distributed Pair Programming

Pair programming (PP) has become popular in the research and software industry as well as being studied for a number of years in computer science courses with positive findings on student performance and attitudes. Advantages of PP reported in the literature are satisfaction, design quality, code productivity, team building, and communication. More recently, distributed pair programming (DPP), which enables two programmers to work remotely, has also attracted the interest of researchers and instructors. The difference between DPP and PP is that the former allows geographically distributed teams to collaborate and share program code. Such collaboration is, thus, only feasible if an underlying infrastructure supports all necessary interactions. The integrated development environments (IDEs) for DPP should cover the basic requirements for remote software development as well as address common PP problems, such as unequal contributions from each member of a pair, feedback during DPP sessions, and communication problems. This paper presents the findings of a study on student performance and attitudes towards DPP in an object-oriented programming (OOP) course. The factors examined were student performance, in terms of assignment grade, exam grade and implementation time in relation to students’ programming experience, and confidence, as well as student attitudes towards DPP, i.e., the feelgood factor, working alone or with a partner, and the perception of their partner’s technical competence. The results suggest that a students’ performance is associated with their programming experience and confidence in programming but not with how comfortable they feel during DPP sessions. Students evaluate the DPP sessions positively regardless of their confidence on programming or their perception of their partners’ technical competence. Students who consider themselves to have about the same programming competence as their partners tend to be more satisfied with DPP sessions. Overall, students prefer working with a partner regardless of their confidence on programming.

Factors Affecting Students’ Performance in Distributed Pair Programming

Pair Programming has been shown to increase productivity and code quality not only in professional software development but also in the context of programming education. The provision of broadband Internet access gave rise to Distributed Pair Programming (DPP) enabling two programmers to collaborate remotely. To gain insight into the benefits of DPP, we performed an empirical study on an object-oriented programming course where 62 students carried out assignments through a DPP platform. The goal of the study is to investigate, in the context of DPP, whether prior programming skills (assessed at the level of student, his or her partner and pair) and pair compatibility are related to student performance. To further examine the effect of DPP on learning outcomes, we have studied whether a pair’s performance on DPP assignments is related to the students’ grade. The findings indicate that the student’s actual skill and the pair’s actual skill affect his or her performance in an object-oriented programming course. The results also suggested that there is no association between pair compatibility and his or her own performance. Finally, pair performance on DPP assignments is related to the individual student performance in the final exams. Such evidence can be used to guide instructors when planning DPP assignments and especially when forming student pairs.

Evaluating the Effects of Scripted Distributed Pair Programming on Student Performance and Participation

The results presented in this paper contribute to research on two different areas of teaching methods: distributed pair programming (DPP) and computer-supported collaborative learning (CSCL). An evaluation study of a DPP system that supports collaboration scripts was conducted over one semester of a computer science course. Seventy-four students participated in the study and used the DPP model to solve programming assignments in pairs, rather than individually. Students were divided into two cohorts in order to examine how best to distribute programming tasks to students, to maximize learning outcomes. The results suggest that while the use of DPP improves pass rates and gets positive feedback from students, the use of collaboration scripts yields equal task contributions from each of the student pair. For a small number of students, overall performance was improved by the adaptive assignment of programming activities, designed to engage students equally in diverse roles and activities.

Distributed Pair Programming Using Collaboration Scripts: An Educational System and Initial Results

Since pair programming appeared in the literature as an effective method of teaching computer programming, many systems were developed to cover the application of pair program- ming over distance. Today's systems serve personal, professional and educational purposes al- lowing distributed teams to work together on the same programming project. The current research focuses in distributed pair programming systems which are suitable for supporting students in learning computer programming. Systematic review of publicly available systems revealed that there is an absence of effective collaboration support for the students. The main drawbacks of pair programming, such as uneven workload distribution and infrequent role switches, cannot be addressed with available systems. While building an enhanced version of a distributed pair programming system, successful instructional strategies in similar collaborative learning sys- tems were explored, in order to improve students' interactions when applying pair programming over distance. As a result, the new system allows students to practice distributed pair program- ming in the form of collaboration scripts. This paper presents the features and the underlying concepts of the system, and the results of its first evaluation. The study showed that distributed pair programming attracted positive feedback from students, and that scripted collaboration af- fected students' engagement in programming, and resulted in an evenly distribution of learning objectives among pairs.

협업을 지원하는 분산 페어 프로그래밍 시스템 설계/구현 및 검증

The dominant trend in software development is the globalisation of the software industry. This development is faced with diverse problems, which require solution by the adoption of new processes and development techniques. eXtreme Programming (known as XP) is one methodology which is now at the leading edge of software development. This recent trend in XP allows organisation members to cooperate towards the development of new software independently of the existing developers. This is achieved functionally between the members by the development of distributed pair programming, this is not IDE plug-in shape of text, simple screen sharing or chatting function based.

Assessing the effectiveness of distributed pair programming for an online informatics curriculum

Studies have shown that distributed pair programming improves student performance and retention in online computer science (CS) courses. However, as online CS courses become more commonly offered in computer science and Informatics departments around the country, it is imperative that distributed pair programming becomes as effective as when performed in co-located spaces such as computer labs. The present study identifies a disparity in student attitudes towards pair programming in co-located versus online environments. This study identifies several qualitative measures that can impact the pedagogical advantages of pair programming when implemented into an existing online computer science curriculum. This on-going study focuses on the online Informatics curriculum at Indiana University Bloomington and Indiana University Purdue University Indianapolis. Begun in Spring 2009, the research focuses on student experiences and perceptions of pair programming, and utilizes both quantitative and qualitative assessment methods. In order to improve the effectiveness of distributed pair programming, it is crucial to properly assess teaching and learning practices that will improve student engagement and motivation in distributed pair programming exercises. Student experience surveys, using a modified Likert scale, demonstrate that student-centered perceptions of the ease and effectiveness of pair programming differs significantly between co-located and online activities. This paper identifies several key areas where there is a noticeable variance between co-located and online pair programming experiences, and argues that addressing and improving these key areas will be vital for the successful implementation and sustainability of distributed pair programming efforts.

Analysis and Design of Distributed Pair Programming System

Pair Programming (PP) that has gained extensive focus within pedagogical and industrial environments is a programming practice in which two programmers use the same computer to work together on analysis, design, and programming of the same segment of code. Distributed Pair Programming (DPP) system is a programming system to aid two programmers, the driver and the navigator, to finish a common task such as analysis, design and programming on the same software from different locations. This paper first reviews the existing DPP tools and discusses the interaction and coordination mechanism in DPP process. By means of activity theory and language-action theory, some basic requirements of the DPP system are presented. Then, a design framework of such system and functions of each sub-system are deeply analyzed. Finally, a system prototype is implemented by plug-in style in Microsoft Visual Studio environment.

Understanding Tools and Practices for Distributed Pair Programming

When considering the principles for eXtreme Programming, distributed eXtreme Programming, especially distributed pair programming, is a paradox predetermined to failure. However, global software development as well as the outsourcing of software development are integral parts of software projects. Hence, the support for distributed pair programming is still a challenging field for tool developers so that failure for distributed pair programming becomes less mandatory. In this paper, we analyse the social interaction in distributed pair programming and investigate how current technology supports this interaction. We present XPairtise, a plug-in for Eclipse that allows instant pair programming in distributed development teams. In addition, we report on experiences and findings when using XPairtise in a distributed software development setting.

Empirical evaluation of distributed pair programming

Pair programming, in which two individuals share a single computer to collaboratively develop software, has been shown to have many benefits in industry and in education. One drawback of pair programming is its collocation requirement, which limits its use to situations where the partners can physically meet. A tool that supported distributed pair programming, in which the partners could pair from separate locations, would remove this impediment. This paper discusses the development and empirical evaluation of such a tool. A significant feature of this tool is the presence of a second cursor that supports gesturing. Students who used the tool in their introductory programming course performed as well as collocated students on their programming assignments and final exam. These students also spent less time working by themselves. They also felt that the gesturing feature was useful and used it regularly.

HOW DISTRIBUTION AFFECTS THE SUCCESS OF PAIR PROGRAMMING

Recent experiments demonstrated the effectiveness of pair programming in terms of quality and productivity. Growing interest towards global software development is fostering the design of suitable methods and tools for distributing software processes, at any level of detail, from entire subprocesses up to a single activity. Consequently, people placed in different locations could also share programming tasks and related practices, such as pair programming. Unfortunately, distribution might seriously compromise the success of pair programming, due to communication and collaboration issues. We have performed an experiment in order to investigate the impact of distribution on pair programming when performing maintenance tasks. An interesting conjecture stems from the experiment: under certain conditions, distributed pair's components tend to dismiss from each other, stopping the collaborative work. This can be a very expensive risk factor to keep into account when planning tasks of distributed pair programming.

Student performance in CS1 with distributed pair programming

Pair programming has been shown to provide many pedagogical benefits, particularly for students who are learning to program. One drawback with pair programming is its collocation requirement, which makes it difficult for some students to enjoy these benefits. We report on an experiment with a tool that allows students to pair program from separate locations. Our results indicate that students who experience the increased flexibility of distributed pairing perform as well in their introductory programming course as students who have to physically meet in order to pair. These students are also as confident as the collocated students. We also confirm earlier results reported by others that pair programming results in greater student confidence and increased student performance.

IDE 기반의 분산 페어 프로그래밍 시스템의 설계 및 구현

In distributed office, the pairs can program together using a distributed pair programming system. Many CSCW tools featuring remote screen sharing function have insufficient usability, accessibility to introduce many developers to distributed pair programming. In this paper, we suggest a distributed pair programming system based on IDE, which many developers will accept and use easily. We have do#eloped a user scenario and GUI of the system, making distributed pair programming easier and designed with high extensibility by adapting design patterns and implemented in Java language. Our findings will be of significant help to developers dealing with implementation of distributed pair programming function into some commercial IDE.