A problem-centred approach to pair programming in the classroom
Whilst I intend to undertake a phenomenological approach to interviewing secondary school teachers about their experiences of maker education in the curriculum for my EdD, I wanted to take a problem-centred focus to interview a primary school teacher about their pedagogical approaches to teaching programming in the classroom. Both qualitative inquiries aim to investigate an emic perspective of maker education and teaching programming as desired outcomes.
My own interest in this area comes from reviewing the relevance of current educational research and shows only a small number of qualitative studies to support classroom teachers’ planning around the phenomenon of pair programming. Technology has been part of planning in schools for many years now but the focus of research has been on the “applications of existing technologies into subject practice” (Passey, D., 2017), and the introduction of a Computing curriculum in 2014 gave another emphasis on computer science with a requirement for every child across key stages one to four to be taught how to program. At this time, over three hundred teachers were surveyed as part of a study at King’s College London into teachers’ challenges and their strategies used to deliver the new subject in primary and secondary settings. Qualitative responses indicated that concerns were mostly about the depth and breadth of teachers’ own Computing subject knowledge, particularly the concepts of computer science and programming. (Sentence and Csizmadia, 2017), and so my focus for a study was to share pedagogical practice and guidance from a practitioner in schools teaching programming with confidence.
There is now a need to increase the number of educational research studies to support teachers in how to adopt new pedagogies such as pair programming, as practitioners make the transition to teach computer science from other subject specialisms. Five years on from the introduction of the curriculum, the Department for Education funded a new National Centre for Computing Education (NCCE) in 2019 as a professional development programme designed to support teachers from non-specialist backgrounds who now teach Computer Science, and my own observations from working at the NCCE support the notion that teachers need and want to further develop strategies to teach Computing.
Located within qualitative research design, the ontological perspective of this study examined pair programming as an approach to teach programming in the primary curriculum within the theme of research into computer science education (Mason, 2002). Pair programming is described as a practice in which “two programmers work collaboratively at one computer, on the same design, algorithm, or code” (Nagappan et al., 2003) and is an industry standard way to code in the workplace which has been adopted by some teachers as an educational approach in schools (Hanks et al., 2011). In the classroom pair programming has been described as two children sharing a computer, taking turns as the driver and navigator and switching between those roles at regular intervals (Braught, Wahls and Eby, 2011), and a quantitative research study by McDowell et. al. (2006) reported that pair programming does improves student retention, confidence, and program quality but the sample of students were aged sixteen years and over as the study followed their progression through undergraduate courses. This pilot gave an opportunity to conduct a qualitative study and interview a primary school teacher who has adopted pair programming over the last two years, contributing and closing the gap with an emic perspective of pedagogical content knowledge and children’s learning of programming using pair programming pedagogy.
Adopting a problem-centred interview methodology helped me to research the challenges facing teachers as I co-constructed knowledge with a practitioner using an interview schedule that facilitated more inductive thinking, thus bringing together an open narrative with more structured questioning throughout the interview. As a pilot I was personally interested in the knowledge shared that can help other teachers and also by the possibility of insights that I hadn’t anticipated having an impact on future research and practice in schools, so I developed the ‘traveller interview’ perspective to my enquiry about pair programming (Kvale and Brinkmann). I was curious to learn through discussion with a flexible direction in interview structure, as such travelling around the landscape of pair programming and computer science education, and recognised that this was a more complex methodology to adopt as an inexperienced researcher. This was in contrast to a more structured approach with the interviewer as a ‘miner’ role, and the metaphor describing how pre-defined information relevant to the research phenomenon is extrapolated during an interview, but as already stated this pilot was an opportunity to test out methods to inform research design of my final EdD thesis. The co-construction of knowledge as data was generated in this study and the extension beyond mining and excavation of data provided greater insights into the topic than I anticipated from a dialogue between me and the teacher as interviewee (Mason, 2015, p.68).
The value of co-production of knowledge with teachers is a fundamental element to my axiological perspective so I used the technique of problem-centred interviews (PCI) to gain first-hand insights of the topic from a teacher using the travelling methodology alongside storytelling and interviewing methods to build an emic perspective of teaching computer science in primary schools. (Witzel and Reiter, 2012).
The interview topic focused on how pair programming as an approach to teaching programming in primary schools, with an overarching research question of:
In what ways can pair programming impact on children’s learning?
Supplementary questions included:
What pedagogical strategies do teachers report work well for teaching computer science in school?
Do teachers find pair programming an effective approach to teach computing?
My design drew on insights from a number of approaches as I combined methods to suit the context of computer science education. Linking practical knowledge through professional practice with research understanding is important to me in order to “bridge the gap between codified research knowledge and the everyday craft knowledge of teachers” (McIntyre, 2005), so for this assignment the research study comprised of one teacher participant based on purposive sampling to provide helpful insights and critical reflection of their adoption of pair programming in a primary school. Using purposeful sampling in this way enabled me to interview the teacher in their natural setting within an hour’s planning and preparation (PPA) time in school and gather narrative accounts of their experiences of teaching computer science using an approach of naturalistic inquiry (Lincoln and Guba 1985).
The interview process consisted of three parts in line with the framework suggested by Kvale and Brinkmann (2009) and this had a number of key advantages, including putting the interviewee at ease which was crucial in my intention to build a dialogue with the teacher and take the traveller interview perspective. I tailored the study by thematizing and designing the methodology with a guided approach through the development of an interview guide. The guide included a list of semi-structured interview questions with examples of probing follow-ups to provide some scaffolding for sequencing and also helped to keep me on track with themes throughout the interview Mason (2005, p.67). Documenting post-interview field notes also added further context during the final data analysis phase. Informed consent was shared beforehand and became a mechanism to promote transparency and honesty throughout the interview study with the inclusion of procedures to increase the credibility of the assignment, ethical values to safeguard the wellbeing and confidentiality of the participant and respect their right to privacy (Creswell & Miller, 2000).
I focused on a fifteen minute section of the interview transcript for data analysis to develop an investigation using Quirkos, an online tool, and this provided me with a first grasp of the phenomena being researched (Silverman, 2015). Adopting grounded theory as a methodology supported me to construct findings from the data as I used inductive reasoning to analyse the teacher’s insights and their perception of the effectiveness of adopting pair programming in the classroom, each time generating codes with subsequent themes for concepts based on the data to contribute to research findings (Glaser and Strauss, 2017).
The themes fell within three broad areas:
Impact - perceived impact of adopting pair programming on engagement and learning
Peer Learning - interactions and behaviours as children work in pairs
Planning and Management - the role of the teacher including instruction and planning
Coding (*) enabled me to use content analysis as a tool to determine, build and analyse sub-groups of data within these analysed themes, all of which were constructed with a coding frame determined by the data collected rather than the research questions. I implemented the five step approach to content analysis from Marvasti (2004) which also supported me to consider the conceptual analysis of the text and consider the reliability of my findings using three criteria from Silverman (2015) of stability, reproducibility and accuracy, recognising eighty per cent as an acceptable margin of error for reliability because of the human nature of researchers and coding errors.
Personal interest in visualising data as objects and through sonification led me to adopt Ball and Smith’s (1992) methods of using ‘research generated’ and ‘found objects’ with the Quirkos software to categorise the data, supporting me as a researcher to count the occurrence of themes and sub-groups and continue the analysis. Research into pair programming can focus on the social-cultural perspective of peer learning in the classroom (Han, Lee and Lee, 2009) or the importance of the compatibility of pairings of students (Williams et al., 2006), and in this instance the teacher concentrated more generally on the role of them as a practitioner to plan and manage a learning environment using pair programming as an approach to effectively teach computing, specifically to build on the progression of programming skills.
*Summarised data and analysis to preserve confidentiality and anonymity
I found the experience of close collaboration between teacher and education researcher to be meaningful as the emic and etic perspectives came to terms with the pedagogical issue of pair programming, and the subsequent co-construction of knowledge was productive and enlightening. The insights and learning through analysis of qualitative data have been beneficial to think about new topics in empirical research such as the fascinating observation about how hardware set-up could contribute to differences in children’s outputs from programming tasks.
Ball, M.S. and Smith, G.W., 1992. Analyzing visual data (Vol. 24). Sage.
Bevan, M.T., 2014. A method of phenomenological interviewing. Qualitative health research, 24(1), pp.136-144.
Braught, G., Wahls, T. and Eby, L.M., 2011. The case for pair programming in the computer science classroom. ACM Transactions on Computing Education (TOCE), 11(1), pp.1-21.
Creswell, J.W. and Miller, D.L., 2000. Determining validity in qualitative inquiry. Theory into practice, 39(3), pp.124-130.
Faja, S., 2014. Evaluating effectiveness of pair programming as a teaching tool in programming courses. Information Systems Education Journal, 12(6), p.36.
Flick, U., 2004. Triangulation in qualitative research. A companion to qualitative research, 3, pp.178-183.
Glaser, B.G. and Strauss, A.L., 2017. Discovery of grounded theory: Strategies for qualitative research. Routledge.
Hammersley, M., 2005. Ethnography and discourse analysis: Incompatible or complementary?. Polifonia, 10(10).
Han, K.W., Lee, E. and Lee, Y., 2009. The impact of a peer-learning agent based on pair programming in a programming course. IEEE Transactions on Education, 53(2), pp.318-327.
Hanks, B., Fitzgerald, S., McCauley, R., Murphy, L. and Zander, C., 2011. Pair programming in education: A literature review. Computer Science Education, 21(2), pp.135-173.
Kvale, S. and Brinkmann, S., 2009. Interviews: Learning the craft of qualitative research interviewing. Sage.
Lincoln, Y.G. and Guba, E.G., EG (1985). Naturalistic inquiry.
Marvasti, A., 2004. Qualitative research in sociology. Sage.
Mason, J., 2002. Designing qualitative research. Qualitative researching, 2.
Mason, J., 2015. Qualitative Researching, London: Sage, Chapter 4, pp. 62-83..
McDowell, C., Werner, L., Bullock, H.E. and Fernald, J., 2006. Pair programming improves student retention, confidence, and program quality. Communications of the ACM, 49(8), pp.90-95.
McIntyre, D., 2005. Bridging the gap between research and practice. Cambridge Journal of education, 35(3), pp.357-382.
Nagappan, N., Williams, L., Ferzli, M., Wiebe, E., Yang, K., Miller, C. and Balik, S., 2003. Improving the CS1 experience with pair programming. ACM SIGCSE Bulletin, 35(1), pp.359-362.
Passey, D., 2017. Computer science (CS) in the compulsory education curriculum: Implications for future research. Education and Information Technologies, 22(2), pp.421-443.
Seidman, I., 2006. Interviewing as qualitative research: A guide for researchers in education and the social sciences. Teachers college press.
Sentance, S. and Csizmadia, A., 2017. Computing in the curriculum: Challenges and strategies from a teacher’s perspective. Education and Information Technologies, 22(2), pp.469-495.
Silverman, D., 2015. Interpreting qualitative data. Sage.
Telford, H., 1996. Transforming schools (Vol. 2). Psychology Press.
Williams, L., Layman, L., Osborne, J. and Katira, N., 2006, July. Examining the compatibility of student pair programmers. In AGILE 2006 (AGILE'06) (pp. 10-pp). IEEE.
Witzel, A. and Reiter, H., 2012. The problem-centred interview. Sage.