Real-time monitoring of dynamic English listening: Introducing the ReMoDEL project

Abstract

Many courses at Swedish universities involve English (Malmstörm & Pecorari, 2022), and the trend towards more English Medium Instruction (EMI) courses is common in other parts of Scandinavia and around the world (e.g., Lasagabaster & Doiz, 2021). However, the amounts of spoken English understood by students in EMI lectures remains unclear (e.g., Ducker, 2022), as previous methods for monitoring listening comprehension are open to criticisms related to memory limitations, time effects, test-construction and so on. To better understand EMI student listening capabilities, to identify specific instances when students report non-comprehension, and to learn about student comprehension strategy use, the ReMoDEL project introduces a novel footpedal mechanism that students use to indicate periods of (non)comprehension.

The project will examine help elucidate:

1.  how much of an EMI lecture students comprehend,
2.  what EMI lecture comprehension difficulties students encounter, and
3. whether and how students overcome said difficulties.

The presentation begins by briefly describing challenges students face when listening during EMI classes and research techniques used in previous EMI listening research. Next, the footpedal mechanism and accompanying research methodology for the ReMoDEL project are introduced, demonstrating a protocol for monitoring second-by-second (non)comprehension triangulated through recordings, stimulated recall interviews, and comprehension tests. This battery of data collection tools allows researchers to monitor the relationship between teachers’ spoken output in EMI lectures and the effects on student comprehension and learning.

The ReMoDEL project will run from 2024-2026 and is funded by the Swedish Research Council. The techniques outlined in the presentation will be used in future data collection for the project and represent a meaningful shift in the possibilities of monitoring comprehension in real time, thereby minimizing weaknesses in previous data collection on listening.