Teaching and Learning English through Corpus-based Approaches in Norwegian Secondary Schools: Identifying Obstacles and a Way Forward

This doctoral dissertation presents the use of corpus-based approaches to English language learning in upper secondary school in Norway. The research was conducted in two distinct phases. The first phase investigated the pedagogic corpus work of four corpus-trained, in service teachers and their students’ corpus literacy alongside factors that might have influenced this work. Data were collected through a questionnaire to students and teacher interviews. The second phase featured a teacher-researcher collaboration with one of the four aforementioned teachers and two of his first-year, upper-secondary classes to design and implement a corpu-based approach over a two-week period. Data were collected through a case-study design with classroom observations, and subsequent student group interviews. Previous studies have shown that corpus-based approaches to language learning result in positive learning outcomes; however, most studies are at the tertiary level and designed and conducted by corpus scholars. Meanwhile, data-driven learning in secondary school is “relatively uncharted territory” (Wicher, 2020) and there has been a call for more qualitative studies. The current dissertation sought to contribute knowledge of data-driven learning in the secondary-school context and insight into the processes and opinions of teachers and learners related to pedagogic corpus use. In the first phase, it was found that the teachers, despite their formal corpus training, had avoided corpus-based approaches in their practice, and few of their students knew anything about corpora. Factors such as teachers’ beliefs about their students’ digital and linguistic competence and about corpora, teachers’ topic focus and epistemic beliefs, and the inaccessibility and cost of corpus applications contributed to their reluctance to introducing their students to corpora. In the second phase, several opportunities for learning were found including instances of metatalk to describe corpus data, peer scaffolding where students helped each other to learn the tool, and teacher scaffolding where the teacher confronted the students with their socio-economic prejudices that arose while working with corpus data from Irish English speakers. However, students’ impressions of the tool and process were negatively skewed. Their critique focused on the absence of the teacher, the complexity and aesthetics of the corpus tool’s interface and data, and the tool’s irrelevance to their learning process. In addition to the empirical contributions described above, it is argued in the dissertation that there are two major obstacles to data-driven learning that need to be addressed in order for its ii application to be normalized in the classroom. These obstacles concern a) the novelty of the approach and the training and mediation required to overcome this novelty, and b) the relevance of the approach to teachers, students, and the curriculum. Inquiry-based education was brought in as a theoretical framework that has considerable overlap with the concepts of data-driven learning but includes a more pronounced social dimension that foster teacher and peer mediation, collaborative learning, and knowledge sharing